Genotypic Characterization of Torymus sinensis (Hymenoptera: Torymidae) After Its Introduction in Tuscany (Italy) for the Biological Control of Dryocosmus kuriphilus (Hymenoptera: Cynipidae).

Torymus sinensis Kamijo (Hymenoptera: Torymidae) is an alien parasitoid that is used in many areas of the world for biological control the Asian chestnut gall wasp, Dryocosmus kuriphilus Yasumatsu (Hymenoptera: Cynipidae). In Italy, this parasitoid was imported from Japan in 2003 and subsequently multiplied and released throughout the country. In this study, a phylogenetic investigation was carried out on insects from three different sites in northern Tuscany (Italy). Moreover, the possible hybridization between T. sinensis and some native Torymus species was evaluated. The conserved region 18S rRNA gene and the hypervariable ITS2 (Internal Transcribed Spacer 2) region of the ribosomal cistrone were selected as molecular markers. Sequencing the amplified products, after cloning, ruled out any hybridization between T. sinensis and the native Torymus species, and also confirmed the presence of two haplotypes for the Tuscan population of T. sinensis both for the region of the 18S rRNA gene as well as for the ITS2 region. These results confirm that the environmental impact of the alien parasitoid T. sinensis in the study site is acceptable, although an extensive and repeated monitoring would be desirable.

Chestnut tree cultivation has played an important role in European silviculture from an economic and environmental point of view (Conedera et al. 2004a, b; Aebi et al. 2007); its relative importance has changed over time and varies across different regions (Conedera et al. 2016). Many phytosanitary threats have affected European chestnut cultivation. The most recent being the Asian chestnut gall wasp (ACGW) Dryocosmus kuriphilus Yasumatsu, which was accidentally introduced from China into Europe in 2002. Already invasive in Korea, Japan, and United States between the 1940s and 1970s (Gibbs et al. 2011), the first European record of ACGWs was in the Piedmont (Italy) in 2002 (Brussino et al. 2002). From there, it spread out rapidly in Italy, and Turkey (Çetin et al. 2014, Avtzis et al. 2018).

The damage caused by ACGWs, whose parthenogenetic females lay their eggs in both leaf and reproductive buds, is due to gall development with consequent severe production losses and a reduction in the photosynthetic surface of the leaves. Resistant chestnut ecotypes or chemical control (EFSA 2010) have been used to manage ACGWs, but only the introduction of an alien parasitoid from China, Torymus sinensis Kamijo (TORYSI) has led to consistent results.

TORYSI is univoltine like its host, but a diapause of 12 mo can occur in some larvae probably because of host deficiency (Ferracini et al. 2015). TORYSI has been introduced successfully into Japan and United States for the biological control of the ACGW (Gibbs et al. 2011). In 2005, TORYSI adults were first released in the Piedmont (Quacchia et al. 2008) and then all over Italy (Quacchia et al. 2014). The first release of TORYSI in Tuscany was in 2010, as part of a regional ACGW biological control project (Conti et al. 2016).

The TORYSI population introduced into Italy was imported from Japan, where was released in 1975 to support the control activity exerted by a native torymid, Torymus beneficus Yasumatsu et Kamijo (Moriya et al. 2003). Later, the molecular marker ribosomal internal transcribed spacers 2 (ITS2) was used by Yara (2006) for discriminating among T. sinensis, T. beneficus, and their hybrids; afterward, Yara et al. (2007) used cytochrome oxidase subunit I (COI), ITS1 and ITS2, demonstrating that in one specimen hybridization had occurred between a female of T. beneficus and a male of TORYSI. This suggests that the hybridization is possible although quite rare. The hybridization of insect native species with introduced natural enemies is a possible impact related to the classical biological control of invasive insects (Van Driesche and Hoddle 2017).

DNA markers are used for assessing genetic diversity, identifying haplotypes, and predicting migration and colonization (Salvato et al. 2002, Llewellyn et al. 2003, Margonari et al. 2004, Bosio et al. 2005, Behura 2006, Guo et al. 2017). Molecular markers are utilized to identify the phylogeny and biogeography of insect populations and to understand the means of evolution and evolutionary trajectories (Luque et al. 2002, Chatterjee and Mohandas 2003, Mohandas et al. 2004, Prasad et al. 2004). The main applications of molecular markers are the following: mating, parentage and kinship, insect plant interaction, insect pathogen interaction, and insect ecology study. Molecular analysis allows a sample to be identified independently of the sex and the stage of the biological cycle. DNA-based techniques have thus proved particularly useful in the study of the taxonomic and phylogenetic relationships of insects (Caterino et al. 2000).

Many molecular markers have been used in various studies and several authors have reviewed the various marker techniques (Lehmann et al. 1997, Kuhner et al. 2000, Black et al. 2001, Nagaraju et al. 2001, Brumfield et al. 2003, Morin et al. 2004).

Ribosomal DNA (rDNA) is the most widely used nuclear sequence in evolutionary analyses. Thanks to its high rate of evolution, the ITS regions flanking the 18S, 5.8S, and 28S regions, ITS1 and ITS2, have been used in phylogenetic inference for closely related taxa (Miller et al. 1996) and phylogeographical and other population genetic studies (Navajas et al. 1998, Ji et al. 2003, Volkov et al. 2003, Long et al. 2004, Mahendran et al. 2006, Yara 2006, Kumar et al. 2018, Li et al. 2018). The conserved region rDNA 18S has been extensively used for evaluating relationships among taxa (Nyaku et al. 2013).

The aim of this paper was to investigate the phylogeny of T. sinensis by comparing ITS2 sequences and conserved region rDNA 18S.

We also investigated whether TORYSI populations in Tuscany (Italy) has undergone hybridization with native species belonging to the same genus.

Materials and Methods

Study Sites

Three sites in northern Tuscany (Italy) were chosen for ACGW gall sampling (Fig. 1): Fosdinovo (FOS), Capezzano Monte (CAPE), and Catagnana-Barga (CATA). Two of these sites, Fosdinovo and Capezzano Monte, face the Mar Ligure coast while Catagnana-Barga is in a valley of the river Serchio. The Apuanian Alps separate Fosdinovo and Capezzano Monte from Catagnana-Barga.

Fig. 1.

Google map of the study sites: FOS = Fosdinovo, CAPE = Capezzano Monte, CATA = Catagnana-Barga.

Insect Collection

In order to collect ACGW parasitoids, about 400 closed galls were collected between 3 March 2016 and 9 March 2016 in each of the three sites.

The galls were split into four plastic containers with a perforated lid for each sampling site and maintained at room temperature and humidity. Twice a week, until mid-May, the containers were monitored to observe parasitoid emergence. The specimens were captured, placed individually into a vial, labeled, and frozen (−20°C). Given that TORYSI adults were the most common specimens, 50 of them were separated from the other parasitoids and labeled according to their origin with the following abbreviations: FOSX (Fosdinovo), CAPEX (Capezzano Monte), CATAX (Catagnana-Barga) where X is a literal and/or numeric code of the isolate. These individuals made up the stock for genetic analysis.

Dichotomous keys (Askew 1961, Graham 1969, Zerova 1978, Vere and Gijswijt 1998, Gibson and Fusu 2016) and comparison with type material available in the Department of Agriculture, Food and Environment (DAFE), University of Pisa and the Department of Agriculture, Food, Environment and Forestry (DAFEF), University of Florence were used to identify parasitoids.

DNA Extraction

Genomic DNA was extracted from individual insects using the Quick-DNA Miniprep Plus Kit (Zymo Research, USA) following the manufacturer’s instructions. The concentration of each DNA sample was measured using a WPA biowave DNA spectrophotometer (Biochrom Ltd., Cambridge, England), and their integrity was evaluated by agarose gel electrophoresis. The DNA was stored at −20°C.

PCR Primer Design

The sequences obtained by amplification with universal primer 18S rRNA gene (Applied Biosystem/Ambion, USA) were aligned using the CLUSTALW program, highlighting the presence of two haplotypes (bands). The amplified DNA sequences were cloned and the two haplotypes were sequenced by automated sequencing (MWG Biotech, Ebersberg, Germany).

The primer pairs for identifying the haplotypes of the partial region of 18S rDNA (To18SA and To18SB), were designed from nucleotide sequences (acc. Nos. MH543348 and MH543349) using Primer3 software (Applied Biosystems) as reported in Table 1. The primers for ITS2 were constructed in homologous regions, after CLUSTALW (Thompson et al. 1994) multialignment of sequences selected by BLASTN analysis of T. sinensis genes for 5.8S rRNA, ITS2, 28S rRNA, partial and complete sequence, isolate: CK15 (acc. no. AB200273); Trichogramma minutum TmMS16 ITS1, 5.8S ribosomal RNA gene, and ITS2, complete sequence (acc. no. AY374440); Leptocybe invasa voucher Li_CN_1 5.8S ribosomal RNA gene, partial sequence; ITS2, complete sequence; and 28S ribosomal RNA gene, partial sequence (acc. no. KP143962); Quadrastichus mendeli 5.8S ribosomal RNA gene, partial sequence; ITS2, complete sequence and 28S ribosomal RNA gene, partial sequence (acc. no. KF879806); and Ooencyrtus pityocampae haplotype 2f 5.8S ribosomal RNA gene, partial sequence; ITS2, complete sequence and 28S ribosomal RNA gene, partial sequence (acc. no. KM527088) (Table 1).

Table 1.

List of Torymus sinensis specific primer sequences used in PCR assays

Primer name	Primer sequence	Gene sequence accession number
ToITS2	F: 5′-TGTGAACTGCAGGACACATG-3′ R: 5′-ATGCTTAAATTYAGCGGGTA-3′	This paper
To18SA	F: 5′-ACTCACCAGGTCCAGACAAAA-3′ R: 5′-GGACATCTAAGGGCATCACAGAC-3′	MH543348
To18SB	F: 5′-CCTCACCAGGCCCGGACACC-3′ R: 5′-GGACATCTAAGGGCATCACAGAC-3′	MH543349

PCR Amplification

Amplification was carried out by conventional polymerase chain reaction (PCR) in 20 μl reactions containing 1× 10× DreamTaq Buffer (Thermo Fisher Scientific, USA) 0.5 μM of each primer (Table 1), 1U of DreamTaq (Thermo Fisher Scientific, USA), and 20 ng of template DNA. PCR was run in a PCR system 2700 (Applied Biosystems, USA): Thermocycling consisted of an initial denaturation step at 95°C (5 min), which was followed by cycles for: Universal primer 18S rRNA gene (Applied Biosystem/Ambion, USA) (95°C for 30 s, 57°C for 30 s, and 72°C for 30 s) 30 cycles; ToITS2 (95°C for 30 s, 50°C for 40 s, and 72°C for 40 s) 40 cycles; To18SA (95°C for 30 s, 54°C for 30 s, and 72°C for 30 s) 30 cycles; To18SB (95°C for 30 s, 56°C for 30 s, and 72°C for 30 s) 30 cycles; final extension step at 72°C (10 min).

All reactions were checked for amplification by gel electrophoresis.

Amplified DNA sequences were directly inserted into a pGEM-T Easy Vector System (Promega, USA). Colony PCR was performed on putatively transformed colonies using M13Forward and M13Reverse as primers. The clones that showed inserts with different molecular weights using gel electrophoresis analyses were sequenced by automated sequencing (MWG Biotech). The sequences were analyzed using BLASTN (https://blast.ncbi.nlm.nih.gov/Blast.cgi) in order to identify them in the GeneBank.

Phylogenetic Analyses

All sequences were multialigned using the CLUSTALW program (https://www.genome.jp/tools-bin/clustalw). Phylogenetic trees were built using the MEGA7 program (Kumar et al. 2016). The evolutionary relationship was estimated based on the statistical model Neighbor-Joining (in MEGA7 program) with a bootstrap number equal to 1,000.

Results

Gall Parasitoids

A total of 1,048 parasitoids emerged from the galls collected in the three sites (Table 2). TORYSI was the most abundant species in the three sampling sites. The other species were polyphagous native parasitoids.

Table 2.

Species and number of parasitoids emerged from the ACGW galls in 2016

Family	Species	Sampling sites			Total
		Fosdinovo (FOS)	Capezzano Monte (CAPE)	Catagnana-Barga (CATA)
Eupelmidae	Eupelmus urozonus Dalm.	3	5	11	19
Eurytomidae	Sycophila biguttata (Swederus)	0	0	1	1
Pteromalidae	Mesopolobus sericeus (Forster)	3	1	8	12
Torymidae	Torymus auratus (Müller)	1	0	1	2
	Torymus flavipes (Walker)	1	0	1	2
	Torymus sinensis Kamijo	105	293	614	1,012

DNA Amplification with the Universal Primer 18S rRNA Gene

DNA of T. sinensis isolates 3 and 14, collected in Fosdinovo (FOS), were amplified with universal primer 18S rRNA (Applied Biosystem/Ambion, USA) gene and two different bands were highlighted, as shown in Fig. 2.

Fig. 2.

Electrophoretic analysis of PCR products of T. sinensis isolates 3 and 14 (fos3, fos14) and isolates kn and ctc (catakn, catactc) with the universal primer 18S rRNA gene.

The PCR products were cloned and, after PCR cloning, each clone containing one or the other band was sequenced. The aligned sequences revealed two haplotypes, A and B, for the partial 18S rDNA gene, as shown in Fig. 3.

Fig. 3.

Alignment of 18S rRNA gene sequences: Haplotype A (ToSA) and Haplotype B (ToSB) of T. sinensis collected in Fosdinovo (FOS).

Phylogenetic analysis of the sequences confirmed the occurrence of two clusters, the ToSA and ToSB haplotypes (Fig. 4).

Fig. 4.

Molecular phylogenetic relationship between partial 18S rDNA haplotypes of sequences of T. sinensis: FOS, collected in Fosdinovo, CAPE, collected in Capezzano Monte, and CATA, collected in Catagnana-Barga. In the molecular phylogenetic relationship, X, Y, and Z represent respectively: the provenance of sample (X), its name (Y), and the clone (Z) (e.g., capesnow1 is provenance ‘cape’, the sample ‘snow’, and the clone ‘1’). The molecular phylogenetic relationship between haplotypes was estimated on the basis of a Neighbor-Joining statistical model. The numbers by the nodes represent the bootstrap support percentage, estimated with 1,000 replicates in MEGA7. In the tree, the branches of two haplotypes (A and B) are highlighted respectively in red and green. The sequence of Bombyx mori was used as an outgroup.

DNA Amplification with the Primer for Internal Transcribed Spacer 2 (ITS2) and Sequences Analysis

DNA amplification products with primer ToITS2 (Table 1), extracted from T. sinensis collected at our three sites, and Torymus flavipes and Torymus auratus from Fosdinovo, were cloned and, after PCR colony screening, clones showing different molecular weights (Fig. 5) were sequenced.

Fig. 5.

Electrophoresis of PCR colony screening of 12 colonies.

After sequence alignment, a phylogenetic tree (Fig. 6) was constructed using all the sequences obtained in this study along with the sequences of T. sinensis, T. auratus and T. flavipes (from specimens collected in the study site), Torymus geranii (species with a Palearctic distribution collected from ACGW galls in several regions of Italy), and T. beneficus (Japanese species). The sequences of T. geranii and T. beneficus were retrieved from public databases. We used the ITS2 of Bombyx mori as the outgroup.

Fig. 6.

Molecular phylogenetic relationship among Internal Transcribed Spacer 2 (ITS2) sequences of T. sinensis: FOS, collected in Fosdinovo, CAPE, collected in Capezzano Monte and CATA, collected in Catagnana-Barga, TA: T. auratus, TB: T. beneficus; TF: T. flavipes, TG: T. geranii, TS: T. sinensis sequence from the gene bank. TA14_2gb|LT821706|, TA14_1gb|LT821705|, TG15Q_1gb|LT821715|, TG12C_1gb|LT821714|, TSgb|LT821666| in the gene bank were from Italy, the remaining specimens came from Japan. In the alignment, X, Y, and Z represent respectively: the provenance of sample (X), its name (Y), and the clone (Z) (e.g., cape1A5 is provenance ‘cape’, the sample ‘1A’, and the clone ‘5’). In the tree, the sequences from the database are highlighted in red, and the branches of two haplotypes C and D are highlighted, respectively in blue and pink. The evolutionary relationship was estimated by the statistical Neighbor-Joining model and the Bootstrap was estimated with 1,000 replications with the MEGA7 program. The sequence of Bombyx mori was used as an outgroup. Asterisks represent a bootstrap of more than 50%.

All sequences isolated in this study belonged to one of two clusters, C and D. All individuals felt into one of the two clusters, except for the capeA isolate, which showed ITS2 sequences from both clusters.

Discussion

The role of T. sinensis as the main parasitoid in ACSW galls was confirmed in the three study sites. However, despite the restricted time frame of the sampling, other native parasitoids also emerged. Many native parasitoids, especially chalcidoidea hymenopterans, have adapted to ACGW larvae in Italy as well as in other sites in Europe where the ACSW was introduced (Quacchia et al. 2008, Matošević and Melika 2013). In fact, they have shifted to the new host and have aided T. sinensis in its role as a biological control agent. However, the parasitism rate of these native parasitoids, which are frequently associated with oak gall wasps, is generally low. The native parasitoids observed in the sampling include species already observed in other sites in Tuscany (Panzavolta et al. 2013, Panzavolta et al. 2018).

For the phylogenetic investigations of T. sinensis, we used nuclear ribosomal DNA (rDNA) given that it is present in many copies in every species and it is known to provide insights into the evolutionary history of different organisms (Nyaku et al. 2013, Costa et al. 2016, Zhang et al. 2017).

While the rRNA genes are conserved among species, the intergenic spacers (ITS1 and 2) evolve rapidly and have been widely used for intraspecific analyses of diversity of numerous organisms, including animals and plants. The conserved region rDNA 18S has been extensively used for evaluating relationships among taxa (Gomulski et al. 2005, Fritz 2006, Nyaku et al. 2013, Venkatesan et al. 2016).

The genetic analysis carried out on some of the TORYSI adults that had emerged from the galls, showed the presence of two haplotypes. No nucleotide difference within each cluster was found with the specific haplotype primers (To18SA and To18SB), differently from Nyaku et al. (2013) who found two variants of the 18S rDNA when they were working on Rotylenchulus reniformis, a plant parasitic nematode.

The phylogenetic analysis using ITS2 sequences showed that the specimens of T. sinensis that we isolated can be differentiated from the two native species of the same genus (T. auratus and T. flavipes) collected in our sampling sites. This analysis also confirms that T. sinensis and T. beneficus belong to the same cluster, confirming the results of Montagna et al. (2018) who analyzed Torymus spp. specimens from different geographic areas, concluding that early-spring T. beneficus individuals are a separate species and that no T. beneficus was imported into Italy to control D. kuriphilus. This analysis also showed that all our isolates belonged to one of the two clusters, except for the capeA isolate which showed ITS2 sequences of both clusters (Supp Fig. S1 [online only]).

These results suggest that, in the area of this study, T. sinensis imported to Italy did not hybridize with the native Torymus species, such as T. auratus and T. flavipes. In fact, none of our isolates share the ITS2 sequence with these species. However, this is also true for T. geranii, whose DNA sequence is in a data bank derived from specimens collected in two regions in the north of Italy (Piedmont and Liguria). The absence until now of hybridization with native species is a positive feature in the evaluation of the environmental impact of TORYSI. However, in order to minimize the environmental risks routine analyses for intentionally introduced natural enemies should be carried out on a larger scale and implemented with other evaluations on behavioral aspects. For instance, the host range of TORYSI was recently demonstrated to be broader than that reported in the literature (Ferracini et al. 2015), since it is attracted by nontarget hosts other than D. kuriphilus.

Conflicts of Interest

The authors declare no conflict of interest.

Click here for additional data file.