Isolation and characterization of 15 microsatellite markers for the highly invasive box tree moth Cydalima perspectalis (Lepidoptera: Crambidae)

In this study, we report the development of a set of 15 polymorphic microsatellite markers for the box tree moth, Cydalima perspectalis (Walker), a highly invasive insect in Europe causing significant damage to natural and ornamental Buxus trees. The markers were characterized for four distant populations in both its native (China, two populations) and invasive ranges (Czech Republic and Turkey, one population each). The number of alleles ranged from 2 to 12. No marker significantly deviated from the Hardy-Weinberg equilibrium for all the populations sampled. These microsatellite markers are promising tools for further studies on the invasive pathways and dispersal pattern of the box tree moth in Europe.


INTRODUCTION
The box tree moth, Cydalima perspectalis (Walker, 1859), is a pest insect native to East Asia. Its natural distribution includes China, Korea and Japan. The larvae feed on leaves and shoots of plants in the genus Buxus, which can result in their death (Wan et al., 2014). This invasive species was fi rst recorded in 2007 in Germany (Krüger, 2008). Since then, it quickly spread across Europe causing signifi cant damage in both urban areas planted with ornamental box trees and natural forest stands of Buxus sempervirens (Kenis et al., 2013). The trade in ornamental box plants between China and Europe is hypothesized as the invasion pathway for the box tree moth (Casteels et al., 2011) and the trade between European countries could account for its fast subsequent spread (Matošević, 2013). Elucidating invasion pathways of invasive species is essential for developing appropriate management programmes and preventing further introductions of Asian populations of C. perspectalis as well as those of other species that may use the same pathways (Estoup & Guillemaud, 2010). Most of our knowledge about invasive pathways comes from historical and observational data, which may be incomplete and/or misleading. In this context, genetic markers have proved to be effi cient tools for disentangling complex invasive pathways. The use of such markers facilitated the detection of a bridgehead effect in the invasion of the harlequin ladybird, Harmonia axyridis, in Europe (Lombaert et al., 2010). A similar approach used in the case of the invasive fruit fl y Drosophila suzukii led to the suggestion that northeast China was the most probable source of the fl y populations invasive in Europe while those invasive in western North America likely originated from southeast China and Hawaii (Fraimout et al., 2017). However, genetic markers Eur. J. Entomol. 115: 264-267, 2018 doi: 10.14411/eje.2018.026

NOTE
A total of 40 candidate primer sets with both an expected size between 90-110 bp, 150-200 bp or 250-300 bp and at least 7 or more repetitions of the microsatellite motif were assessed for amplifi cation performance in 8 individuals of C. perspectalis from native and invaded ranges. A total of 15 primer sets were unlikely to provide a clear amplifi cation pattern. Then, a selection of sequences with target microsatellites were used to design primer sets using BLAST, ClustalW and Primer3 programmes. Among 4,243 sequences containing a microsatellite motif, 336 primer sets were designed using this software.  selected and microsatellite forward primers were labelled using fl uorescent dyes NED, 6-FAM, PET and VIC (Applied Biosystem, Renfrewshire, UK) to allow for PCR product multiplexing (Table 2). PCR amplifi cations were done in a total volume of 10 μL using the Multiplex PCR Kit (Qiagen®). Thermocycling was carried out in a Veriti® 96 well fast thermal cycler (Applied Bio-systems®) and the following protocol: initial denaturation step of 94°C for 15 min followed by 40 amplifi cation cycles of 94°C for 30 s, 58-61°C for 45 s, 72°C for 30 s and a fi nal elongation step at 72°C for 30 min. Fragments were run on an ABI Prism 3500 sequencer. Raw data for each fl uorescent DNA product were visualized using GeneMapper v. 4.1 (Applied Biosystems®) and allele sizes were scored against an internal GenScan-600 LIZ® Size Standard (Applied Biosystem®).
The fi fteen microsatellite markers were characterized using 4 populations (29-30 individuals each): 2 populations sampled in the two distant parts of the native range of the box tree moth in China (southwestern China -Lijiang and southeastern China -Fuyang) and 2 populations sampled in two distant parts in the area presently invaded in Europe and Asia Minor (Czech Republic -Brno and Turkey -Istanbul; Table 1). Observed and expected heterozygosity, deviation from Hardy-Weinberg proportion (HW) and Linkage Disequilibrium (LD) were calculated using Arlequin v. 3.5 (Excoffi er & Lischer, 2010), and the conservative correction of Bonferroni was not applied. The presence of null alleles was checked using FreeNA (Chapuis & Estoup, 2007).

RESULTS AND DISCUSSION
The total number of alleles ranged from 2 to 12 (Table 2). Two loci (BTM01 and BTM34) were found to be monomorphic in the invading populations studied (Table 3) whereas in the native populations BTM01 was monomorphic for those at Lijiang and BTM34 monomorphic for those at Fuyang. Observed and expected heterozygosity ranged from 0.000 to 0.900 and 0.066 to 0.858, respectively. Global expected heterozygosity He was 0.518 at Fuyang, 0.519 at Lijiang, 0.386 at Brno and 0.441 at Istanbul. Signifi cant departures from HW proportion were detected for different markers in the four populations sampled. Nevertheless, no primer pair signifi cantly deviated from HW in the four populations. LD results showed that only one pair (BTM13-BTM15) was linked in the populations from Fuyang, Brno and Istanbul. The other microsatellite markers did not show signifi cant LD in the four populations. Null allele frequency ranged from 0.000 to 0.237. The null allele frequency of six markers was larger than 15% in at least one population (BTM01, BTM06, BTM13, BTM15, BTM16 and BTM22). Presence of null alleles and HW disequilibrium for some loci was not surprising since it has also been observed in microsatellite sets developed for other lepidopteran species (Kim et al., 2008;Sinama et al., 2011;Lebigre et al., 2015). Moreover, a part of the HW disequilibrium recorded for BTM01, BTM06, BTM13, BTM15, BTM16 and BTM22 loci in no more than three of the four populations studied could be explained by the presence of null alleles, sex-linked loci or population demography history.
For further genetic analyses on C. perspectalis, sex-linked disequilibrium still needs to be investigated since we did not have enough adults in the genotyped populations to test for it. In addition, BTM01 loci characteristics will need to be tested in other populations to determine the utility of this marker because this primer was found to be monomorphic in three of the four populations studied, and signifi cantly deviated from HWE and there was a high null allele frequency in the Fuyang population. Moreover, LD, especially in the native area, needs to be more investigated for the BTM13 and BTM15 markers. Nevertheless, these fi rst microsatellite markers available for C. perspectalis are promising molecular tools for further analysing the invasion routes of this highly invasive pest and disentangling the dispersal patterns that could explain its very fast spread in Europe and the Caucasus region.  Information for each population presented includes Number of observed alleles (Na); Observed heterozygosity (Ho); Expected heterozygosity (He); and Null allele frequency (f). P-value of the deviation from Hardy-Weinberg proportion (HW) provided (without Bonferroni correction).