New internal primers targeting short fragments of the mitochondrial COI region for archival specimens from the subfamily Aphidiinae ( Hymenoptera , Braconidae )

Archival specimens are a great resource for molecular research in population biology, taxonomy and conservation. A primary goal for researchers is to preserve specimens from collections by improving noninvasive methods for DNA extraction and to achieve successful amplification of the short fragments of a target gene in the event of DNA fragmentation. We tested the suitability of a noninvasive method of DNA extraction and amplification of the barcoding region of the mitochondrial gene cytochrome c oxidase subunit I from archival specimens of aphid parasitoids belonging to the genera Aphidius, Lysiphlebus and Praon (Aphidiinae, Braconidae, Hymenoptera). Using a commercial kit as a noninvasive method, we successfully extracted DNA from dry 7 to 41 year old samples of 26 different parasitoid species. However, amplification of the barcoding region failed using the standard primer pair LCO1490/HCO2198. In order to reconstruct DNA barcodes we designed internal genus-specific degenerative primers and a new amplification protocol to target the short fragments within the mitochondrial region. Novel primers were designed using as a template the reference sequences from congeners retrieved from the public database. The combination of standard primers with internal primers, in direct and nested amplification reactions, produced short overlapping subsequences, concatenated to recover long barcoding sequences. Additional analyses also confirmed that primers initially designed for Aphidius, Lysiphlebus and Praon can be combined in a mixture, and successfully used to obtain short fragments of disintegrated DNA from archival specimens of several other braconid species from the genera Ephedrus and Monoctonus. JHR 64: 191–210 (2018) doi: 10.3897/jhr.64.25399 http://jhr.pensoft.net Copyright Milana Mitrović, Željko Tomanović. This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. RESEARCH ARTICLE M. Mitrović & Ž. Tomanović / Journal of Hymenoptera Research 64: 191–210 (2018) 192


Introduction
The DNA from an archival species is an important source of data in the areas of population genetics, conservation, taxonomy and phylogeny.In the past researchers were in conflict between the maintenance of specimens undamaged and their use in molecular analyses, which created a strong limitation for studies on museum specimens, in particular studies with rare or extinct species, or those restricted to one or a few individuals collected many years ago (Gilbert et al. 2007;Mandrioli 2008).However, archival DNA study is now a rapidly developing area of research due to the continual improvements of molecular tools with which it is possible to recover DNA information from museum specimens and dry remains, without damaging the material.
Insects are a group where these tools have received increasing attention and noninvasive techniques have been developed and used for a variety of orders (Gilbert et al. 2007;Andersen and Mills 2012).Noninvasive methods of DNA extraction from dried specimens are important in order to preserve the quality of museum specimens.Unfortunately, not all specimens contain DNA of suitable quality and in the right amount for conclusive genetic studies.Successful amplification depends on post-mortem processes of DNA degradation, which can cause miscoding lesions or physical destruction of the DNA molecule (Rizzi et al. 2012).Degradation of DNA consequently produces methodological difficulties in amplification and sequencing of the target region, processes that are limited by the small quantity of template DNA and recovery of short fragments.Besides natural processes of disintegration, another factor that makes archival specimens difficult to work with is the preservation methodology, which can over time result in DNA damage (Dillon et al. 1996;Burrell et al. 2015).In the case of parasitic Hymenoptera, Andersen and Mills (2012) determined that age was a significant factor for successful sequencing, while size and DNA concentration did not influence the amplification of the targeted nuclear and mitochondrial genes.
Parasitoid Hymenoptera are a taxonomically challenging group under frequent revision, making them a group of great interest for retrieval of genetic information from museum specimens (Andersen and Mills 2012).Among parasitoids that have been intensively surveyed by taxonomists and ecologists are aphid endoparasitoids from the subfamily Aphidiinae (Braconidae, Hymenoptera).They are distributed worldwide, closely following the distribution of their aphid hosts (Starý 1988).As solitary endoparasitoids, Aphidiinae are one of the most important natural enemies of aphids and can effectively regulate their populations (Hågvar and Hofsvang 1991).They have been commercially produced and released as classical biological control agents of aphids in many regions and have achieved significant results in diverse agroecosystems.The most important genera of aphid parasitoids used in biological control are Aphidius Nees, 1818; Diaeretiella Starý, 1960;Ephedrus Haliday, 1833and Praon Haliday, 1833(Boivin et al. 2012).
The subfamily Aphidiinae is a diverse group with many cryptic species complexes, and reliable identification is therefore of key importance for their use as biological control agents.
This study included aphid parasitoids belonging to the common aphidiine genera Aphidius, Lysiphlebus Förster, 1862 and Praon.Identification based on morphology has often been shown to be inadequate in distinguishing the species of these genera due to the limited number of valid discriminatory morphological characters, as well as their high variation on the intraspecific level (Pungerl 1983;Kavallieratos et al. 2005Kavallieratos et al. , 2010;;Tomanović et al. 2003Tomanović et al. , 2004)).Furthermore, several species have confusing taxonomic histories and are in need of revision.In fact, over the last two decades these genera have been constantly rearranged on the basis of new morphological characters and more recently obtained molecular data as well.
Considering that these parasitoids are important for fundamental taxonomic and conservation research, as well as being potential biological control agents in aphid management programs, it would be of great value to investigate the possibility of recovering barcoding fragments of COI from museum specimens.Thus, the main objectives of this study were as follows: i) DNA extraction from dry archival specimens belonging to the genera Aphidius, Lysiphlebus and Praon using a noninvasive method; ii) PCR amplification of several short and overlapping fragments within the barcoding region of cytochrome c oxidase subunit I, iii) traditional Sanger sequencing and alignment of different short overlapping fragments and concatenation to recover longer target barcoding region of mitochondrial DNA and iv) testing the suitability of novel primers for targeting barcodes in archival specimens of other braconid species.

Material and methods
Analyses included species from three different genera of aphid parasitoids, viz., Aphidius, Praon and Lysiphlebus.In total 45 specimens were submitted to molecular analyses, including 11 species of Aphidius, nine of Lysiphlebus and six of Praon, killed and preserved in dry condition from 7 to 41 years prior to DNA extraction (Table 1).Additionally, in order to test the suitability of these primers in amplification of other parasitiods we chose four species from the genus Monoctonus Haliday, 1833 and four of Ephedrus Haliday, 1833, all dry material up to 31 year old (Table 1).

DNA extraction
Dry specimens were carefully removed from the card points so that they could be remounted afterwards if the specimens are holotypes.The whole specimens were used for DNA extraction using the QIAGEN Dneasy Blood and Tissue Kit.In the case of parasitoid specimens used as a control, they were preserved in 96% ethanol prior to extraction.Whole specimens were placed in 2 ml Eppendorf tubes with proteinase K and ATL buffer.After incubation overnight at 56 °C insect specimens were removed from the buffer, rinsed with 96% ethanol several times, air-dried and put back in the collection.The remaining solution was treated according to the manufacturer's instructions.

PCR amplification
The first step was an attempt to amplify a barcoding region of mitochondrial gene cytochrome c oxidase subunit I from dry material using the standard primer pair LCO1490/HCO2198 (Folmer et al. 1994).Each PCR reaction was carried out in a volume of 20μl, including: 1μl of extracted DNA, 11.8 μl H 2 0, 2 μl High Yield Reac-tion Buffer A with 1xMg, 1.8 μl of MgCl 2 2.25 mM, 1.2 μl of dNTP 0.6 mM, 1μl LCO1490 0.5 μM, 1μl HCO2198 0.5 μM, 0.2 μl DNA polymerase 0.05U/μl.The amplification protocol included : i) initial denaturation at 95 °C for 5 min; ii) 35 cycles of 1 min at 94 °C, 1 min at 54 °C and 30 sec at 72 °C and iii) final extension at 72 °C for 7 min.Products were visualized on agarose gel.Due to DNA fragmentation in dry specimens, internal degenerative primers were designed to amplify overlaping short fragments of COI through direct and nested PCR, which could thereafter be aligned to a longer barcoding sequence (Fig. 1).Reference COI sequences of parasitoids retrieved from the GenBank (www.ncbi.nlm.nih.gov/Genbank) were used as a template to design primers for dry material of the genera Aphidius, Praon and Lysiphlebus (Table 2).They were aligned and manually searched for shared conservative regions on which to place the newly designed primers.
The initial idea was to divide the barcoding fragment of COI obtained with LCO1490/HCO2198 into three overlapping subsequences, around 260 bp, 270 bp and 280 bp long respectively, and the primers designed for this were marked as for direct PCR.Furthermore, additional internal primers were designed within these three subsequences to amplify even shorter fragments through nested PCR (Fig. 1).
The genus-specific degenerative primers were used in combination with standard primers LCO1490 and HCO2198 (Fig. 1).Finally, the position of internal primers allowed diverse combinations and targeting of overlapping fragments of different length and position.Due to the shared conservative sites in COI sequences, it was possible for primers initially designed for Aphidius species to be also used in amplification of short fragments in combination with primers specifically designed for Lysiphlebus species (Aph1Rn, Aph2Fd, Aph3Rn) and for dry Praon specimens as well (Aph2Fn) (Fig. 1).
After confirmation of their suitability, the new primers were then used in trials with dry specimens.Products of PCR were obtained in 40 μl volumes.In the direct PCR reac-tion, 4 μl of extracted DNA was added into 36 μl of mix, following the recipe described for the LCO1490/HCO2198 primer pair.In nested PCR, 0.25 μl of a product from direct PCR was added into 39.75 μl of mix.The following protocol was developed for direct and nested PCR: i) initial denaturation at 95 °C for 5 min; ii) 37 cycles of 1 min at 95 °C, 1 min at 54 °C, and 30 sec at 72 °C; and iii) final extension at 72 °C for 7 min.
Amplified COI fragments were sequenced in both directions using an automated equipment (Macrogen Inc, Seoul, South Korea).Overlapping short fragments of the barcoding region were manually edited in FINCHTV ver.1.4.0 (www.geospiza.com),concatenated to obtain longer sequences and aligned using the CLUSTAL W program integrated in MEGA5 (Tamura et al. 2011).A Maximum likelihood tree was constructed using the MEGA5 software, with 500 bootstrap replicates performed to assess the branch support.The evolutionary distances were computed using the Tamura-Nei  method (Tamura and Nei, 1993).Phylogenetic analyses included the sequenced barcodes recovered from archival parasitoid specimens combined with the reference COI sequences of Aphidiinae from GenBank.

Results
Initial trials with dry specimens using standard primer pair for the COI barcoding region LCO1490/HCO2198 failed to give products.Starý, 1960;A. funebris Mackauer, 1961;A. ervi Haliday, 1834;A. smithi Subba Rao and Sharma, 1959;A. sussi) were submitted to molecular analyses (Table 1).Insects had been killed and stored dry in collections for 8 to 41 years prior to DNA extraction.The same three combinations of standard and degenerative primers previously confirmed as suitable in the test trials with fresh material were used with dry samples AD1-AD15 as well.Direct PCR produced amplicons in all three combinations for samples AD1 to AD6, while in the cases of samples AD7 to AD15 no product was visualized.The products from direct PCR with primer pair LCO1490/Aph1Rd were submitted to two independent nested reactions with primers LCO1490/Aph1Rn and Aph1Fn/Aph1Rd; from direct PCR with primers Aph2Fd/Aph2Rd to nested reactions with Aph2Fd/Aph2Rn and Aph2Fn/Aph2Rd; and products obtained with Aph3Fd/HCO2198 were included in nested trials with the primers Aph3Fd/Aph3Rn and Aph3Fn/HCO2198.In all six individual nested reactions short fragments of the barcoding region were amplified successfully and visualized for all of the tested samples.In total 15 specimens of eight Praon species preserved dry for 7 to 20 years prior to DNA extraction were analysed (Table 1).We attempted to retrieve short overlapping fragments of COI barcodes from dry samples PD1-PD15 through the same three direct amplifications as with the fresh material.In analyses with primers targeting the first fragment of the barcoding sequence, all products were obtained.In the second and third reactions short fragments of barcode were amplified in samples PD1-PD11 and PD13, while no product was visualized for samples PD12, PD14 and PD15.The same methodological approach was applied here, namely using the products from direct PCR as a template for secondary nested trials.The amplicons of samples PD12, PD14 and PD15 from the trial with primer pair Pr2Fd/Pr2Rd were processed further in two nested reactions with combinations Pr2Fd/Pr2Rn and Aph2Fn/Pr2Rd, while the products of direct PCR with Pr3Fd/ HCO2198 were processed in secondary analyses using the combinations Pr3Fd/Pr3Rn and Pr3Fn/HCO2198.Subsequent analyses successfully targeted short fragments within the subsequences of the barcoding region in all four nested test trials (Fig. 4).
The novel primers were tested on Lysiphlebus alpinus Starý, 1971;L. confusus Tremblay & Eady, 1978;L. desertorum Starý, 1965;L. fabarum;L. fritzmuelleri Mackauer, 1960;L. hirticornis;L. melandriicola Starý, 1961;L. testaceipes), stored dry in collections for 7 to 22 years.Three separate analyses were conducted using the primer combinations confirmed as suitable with fresh material.Amplicons were visualized in the first direct analysis with the LCO1490/Lys1Rd combination for samples LD1-LD7 and LD10-LD15.No products were visible for samples LD8 and LD9 which were further processed in nested trials with LCO1490/Lys1Rn and Lys1Fn/Lys1Rd.Products of the direct PCR conducted with the primer combination Aph2Fd/Lys2Rd were obtained in all samples except LD8, LD9 and LD12 which were thereafter processed in nested analyses with 1. Aph2Fd/Lys2Rn; and 2. Lys2Fn/Lys2Rd.In the third direct  PCR trial, amplicons were visualized in all analyzed specimens besides LD8, LD9 and LD13 which were further submitted to analyses with primers 1. Lys3Fd/Aph3Rn; and 2. Lys3Fn/HCO2198.We obtained products in all nested trials (Fig. 6).
Our research covers different taxonomically challenging Aphidiinae, for which reason we tested suitability of the newly designed primers on several other archival specimens from the genera Monoctonus and Ephedrus.In order to preserve the limited amount of DNA obtained from dry specimens, we avoided blind PCR trials as well as testing of all possible combinations by doing initial alignment of barcode sequences of fresh material (unpublished data) and degenerative primers (Table 4).According to the alignment we chose the primers best suited to target the species of interest.In the case of Ephedrus species, we chose two combinations for direct PCR, i.e., 1. LCO1490/Pr2Rd, and 2. Aph3Fd/HCO2198.Four species preserved in dry condition for 14 to 31 years in collections were included in the test trials, viz., E. plagiator Nees, 1811 (ED1); E. laevicollis Thomson, 1895 (ED2); E. validus Haliday, 1833 (ED3); and E. koponeni Halme, 1992 (ED4) (Table 1).Amplicons of both targeted fragments were visualized on gel for specimens ED1, ED3, and ED4, while in the case of the ED2 sample a PCR product was visible only with primer pair Aph3Fd/ HCO2198.Products of the ED2 were subjected to separate nested reactions with primer pair LCO1490/Pr1Rd and Pr2Fd/Pr2Rd.Both short fragments of the barcode were successfully amplified and concatenated with the third subsequence obtained in direct PCR to retrieve a longer barcode fragment of COI.
Dry specimens of the following four Monoctonus species preserved for 13 to 26 years were subjected to PCR analyses: M. paulensis (Ashmead) (MD1); M. allisoni Pike and Starý, 2003 (MD2); M. washingtonensis Pike and Starý, 1995 (MD3); and M. leclanthi Tomanović and Starý, 2002 (MD4).The same approach was repeated as with Ephedrus, i.e., barcoding sequences of fresh material were aligned and analysed for primers suitability prior to molecular analyses (Table 4).The final choice fell on three combinations in direct PCR to retrieve three overlapping short fragments within the barcoding COI fragment: 1. LCO1490/Aph1Rd; 2. Pr2Fd/Lys2Rd; and 3. Pr3Fd/HCO2198.The final results show that the tested combinations of standard and degenerative primers successfully amplified all three short subsequences in all tested Monoctonus species.
The overall results of combining different primers in direct and secondary nested reactions are summarized in Fig. 7.

Discussion and conclusion
The barcoding method has shown to be a useful tool in discriminating parasitoid species from the five Aphidiinae genera studied, enabling further research on their biodiversity and phylogeny.The results presented here indicate the possibility of testing many other different combinations of primers in future research on archival specimens with the expectation of achieving success in retrieving the targeted subsequences.The position of the newly designed primers was evidently well chosen, targeting sites conservative enough to permit their multiple uses on a much wider spectrum of museum material than initially planned.
Similar to the results obtained by Andersen and Mills (2012), in our study age was apparently a limiting factor for successful amplification with the newly designed internal primers.On the other hand, the starting point in this study was awareness that museum specimens are not always available, or that the type material is sometimes restricted to a single specimen, etc., and thus cannot be manipulated in numerous trials.For this reason, blank PCR products were always further processed through secondary analyses with additional internal primers.This assumption was confirmed to be the basis of a good methodological approach with substantial success.
The results presented above refer only to combination of primers randomly selected to test their suitability in retrieving the barcoding region from Ephedrus and Monoctonus species.Without the need for further expenditure of limited DNA sources, the here presented overview of nucleotide differences between the barcodes of parasitoids and information about primers clearly indicate that quite a few other combinations can be tested with the expectation of successfully retrieving short fragments.
Many benefits of using novel primers in conservation genetics and phylogeny studies are recognized, above all, the possibility of analyzing archival material of Aphidiinae parasitoids with unresolved taxonomic status.To date there have been many phylogenetic studies with different hypotheses about the origin and classification of certain taxa.Many examples in the literature show the importance of an integrative approach combining molecular and morphological data in taxonomic, phylogenetic and conservation studies, but even when using such an approach, researchers are quite often left with open questions.In view of the many confronting opinions held by different groups of authors, we can assume that the involvement of archival remains of Aphidiinae in molecular analyses will prove to be of great usefulness by yielding results enabling us to resolve the problems of phylogenetic relationships and the taxonomic recognition of different parasitoid groups.
It can be predicted that the herein described method of retrieving the barcoding region in parasitoids will take on increasing importance by making it possible to include not only extinct species preserved in museums, but also endemic or rare species under threat of extinction as well.Good examples of parasitoid species with potential risk of extinction are various associations of aphid hosts/parasitoids whose distribution are restricted to habitats under constant anthropogenic pressure of degradation such as the wetlands (Tomanović et al. 2012).
Modern genomic research opened complex questions exceeding the capacity of traditional DNA sequencing technologies.The Next-generation sequencing has revolutionized the biological sciences allowing us to study biological systems at higher level.In the light of an ongoing rapid progress in the field of modern sequencing technologies, newly designed primers could meet the demands in terms of depth of information in studying genomics of different Aphidiinae by delivering an insight into DNA variation of the target mitochondrial region.

Figure 4 .
Figure 4. Agarose gel visualizing the products of nested trials with products of direct PCR for samples PD12 -P.barbatum, PD14 -P.yomenae, and PD15 -P.yomenae.The products from PCR with Pr2Fd/ Pr2Rd were submitted to secondary nested trials with primer pairs Pr2Fd/Pr2Rn and Aph2Fn/Pr2Rd.Amplicons obtained with Pr3Fd/HCO2198 were used as the template for nested reactions with Pr3Fd/ Pr3Rn and Pr3Fn/HCO2198.

Figure 7 .
Figure 7. Scheme with overview of PCR attempts to recover the barcoding region of cytochrome c oxidase subunit I with novel primers from archival specimens from the genera Aphidius, Praon, Lysiphlebus, Ephedrus and Monoctonus.Primer pairs coloured red were used in direct PCR; black coloured primers were used in secondary nested reactions.Positions where short fragments within the subsequences overlap are marked with a pattern.

Figure 8 .
Figure8.The evolutionary history was inferred by using the Maximum Likelihood method based on the Tamura-Nei model.The tree with the highest log likelihood is shown.There were a total of 568 positions in the final dataset.Initial tree(s) for the heuristic search were obtained automatically by applying Neighbor-Join and BioNJ algorithms to a matrix of pairwise distances estimated using the Maximum Composite Likelihood (MCL) approach, and then selecting the topology with superior log likelihood value.The tree is drawn to scale, with branch lengths measured in the number of substitutions per site.The percentage of replicate trees >50% in which the associated taxa clustered together in the bootstrap test (500 replicates) are shown next to the branches.

Table 1 .
The list of analyzed species from the genera Aphidius, Lysiphlebus, Praon, Ephedrus, Monoctonus with designated aphid host/plant associations and geographic origin.

Sample code Parasitoid species Country of origin Sampling year/ age of samples* Host plant Aphid host
* number of years the specimens were kept dry in collections prior to DNA extraction **Specimen condition: (F) fresh refers to specimens kept after collection in 96% ethanol; (D) dry are specimens which were kept dry in collections, pinned or glued to cardboard

Table 3 .
The list of primers designed for the genera Aphidius, Lysiphlebus and Praon to amplify short fragments of COI barcoding region from dry specimens through direct and nested PCR analyses.
*the last letter in the primer's name refers to PCR reaction: d-direct and n-nested **degenerative base designation/actual base coded: R or -A, or -G; Y or -C or -T; W or -A, or -T.

Table 4 .
Comparison of barcode fragments of COI for Monoctonus and Ephedrus with degenerative primers sequences.