Vouchering of Forensically Important Fly Specimens by Nondestructive DNA Extraction

DNA extraction frequently requires destruction of whole samples. However, when the sample is very rare or has taxonomic importance, nondestructive DNA extraction is required for preservation of voucher specimens. In the case of arthropod specimens, minor anatomical structures such as a single leg or a single wing are oen sacri�ced instead of the whole body for DNA extraction. In an attempt to save the entire anatomical structure of specimens, several authors tried to brew the whole specimen in a lysis buffer and to extract DNA from the “soup.” �e applied this nondestructive DNA extraction technique to a forensically important blow�y species, Phaenicia sericata. �ith nondestructive DNA extraction, a satisfactory quantity and quality of DNA for PCR ampli�cation was obtained with only minimal anatomical disruptions that do not alter the morphologic identi�cation. is nondestructive method may be applicable to DNA extraction of rare samples as well as vouchering of regular �y samples.


Introduction
Most DNA extraction methods inevitably consume at least a small portion of biological specimens.Of course, minor damages to samples do not matter in many instances and many small insect samples are even totally grinded for DNA extraction [1][2][3][4].However, because photographic records prior to sample destruction do not always preserve all the important morphologic features, saving voucher specimens is advisable for specimens with taxonomic importance.In the case of forensically important �y species, many authors have utilized small portions of the body such as legs, wings, thoracic muscle, and entire thorax [5][6][7][8][9].However, preserving the entirety of the specimen is oen preferred, and dissection of speci�c anatomical parts such as the thoracic muscle is oen complex and time consuming.In 1995, Phillips and Simon tried to extract DNA from unimpaired arthropod specimens in the museum collection.However, Phillips and Simon made multiple punctures to the exoskeletons prior to the extraction and did not describe the detailed morphological disruptions aer the extraction [10].e idea to brew whole unimpaired samples in a lysis buffer and to extract DNA from the "soup" has emerged in the �eld of zoology.Rohland et al. extracted DNA from mammalian teeth without any destruction of the specimen by incubating the whole specimen in a digestion buffer [11].Gilbert et al. and Rowley et al. applied Rohland's method to beetles and various insect specimens and tested not only the DNA yields and qualities but also degrees of damages to the exoskeletons [12,13].Although Rowley et al. included a �y species Delphinia picta in their study, it is taxonomically distant from most forensically important �y species [13].To our knowledge, there has been no study which applied the nondestructive DNA extraction technique to forensically important �y species.�e tested a

PCR Ampli�cation.
All �ve samples displayed visible DNA bands for all three primer pairs (Figure 1).However, band intensities were weak for the primer pair COI-F(tRNA Cystein) and COI-R(COII), for which the amplicon size is 1738 base pairs including the whole mitochondrial COI gene.In the cases of shorter amplicons, both mitochondrial (629 base pairs) and nuclear (492 base pairs) markers were excellently ampli�ed.
�.�.Di�r��tion o� �e� �eat�re� �or �or��o�o�ica� I�enti�ca� tion aer DNA Extraction.e overall morphologic features were excellently preserved aer DNA extraction in all samples.Generally, the metallic green gloss which is typical for �uciliinae �y species became slightly dull (Figure 2).A few slight disruptions to particular morphologic features were observed.Although even small bristles were not disrupted, one of the two mid legs was lost in two samples during DNA extraction (Figure 3).Despite shrinkage to the membranous part and a few disruptions of veins, the general integrity of the wings remained intact and suitable for identi�cation of venous branching patterns (Figure 4).Disruptions in morphological features aer DNA extraction are summarized in Table 3.

Discussion
e numbers of nucleotide sequences of forensically important �y species in NCBI Genbank are rapidly increasing.However, sequence acceptance in NCBI Genbank does not guarantee the accuracy of species identi�cation of the specimen.In the �eld of forensic entomology, several sequences of Calliphoridae �y species, which are quite di�erent from other conspeci�c sequences or identical to heterospeci�c sequences, hamper accurate species identi�cation using DNA sequences [15,16].erefore, in those cases, it is ideal for sequence submitters to share their voucher specimens for recon�rmation of morphological identi�cation keys.According to our results using nondestructive DNA extraction, all of the �ve samples did not show any signi�cant morphologic disruption which may affect species identi�cation.Moreover, nondestructive DNA extraction provides a satisfactory quantity and quality of DNA for PCR ampli�cation.�is nondestructive method may be applicable to DNA extraction of rare samples as well as vouchering of regular �y samples.Although we applied traditional phenol/chloroform technique in this study, the incubated "soup" may be transferred to commercially available DNA extraction kits using spin columns.

F 1 :F 2 :F 3 :T 3 :F 4 :
Electrophoresis results for 1738 and 629 base pair amplicons of COI genes (a and b, resp.) and 492 base pair amplicons of ITS2 region (c)."M" stands for "molecular size marker." �omparing dorsal �ie�s of a �y sample before and a�er �N� e�traction, the o�erall morphology is �ell preser�ed e�cept for a slight dullness of metallic gloss and disarray of dorsal setae.�ateral �ie� of a �y sample of �hich the le� mid leg has been lost (arro�) a�er �N� e�traction.Key morphological features aer DNA extraction.Right wing of a �y sample before and aer DNA extraction shows generalized shrinkage and two disruptions in the costal vein aer DNA extraction.

Table 1 .
For each 10 L reaction volume, 20 ng of �y DNA, 25 nmole of MgCl 2 , 2 pmole of each dNTP, 2 × 10 −6 pmole of each primer, 0.2 U of Amplitaq Gold DNA polymerase (Applied Biosystems, Foster City, CA), and 1.0 L of 10x PCR buffer were used.Aer 1 cycle at 95 ∘ C for 11 min of hot start enzyme activation, �y DNAs were ampli�ed by 35 cycles at 95 ∘ C for 30 sec, 55 ∘ C for 30 sec, 72 ∘ C for 90 sec, and 72 ∘ C for 15 min of �nal extension with [13]nicia sericata (P.sericata) specimens and compared key morphological features before and aer DNA extraction.2.Materials and Methods2.1.Fly Specimens.Five arti�cially reared individuals of P. sericata were collected from the cage.Collected �ies were anesthetized with ethylacetate and then submerged in 70% ethyl alcohol solution.Aer overnight treatment at −20 ∘ C, submerged specimens were picked out and air-dried at room temperature.Dried specimens were observed and recorded as digital images under a dissecting microscope.2.2.DNA Extraction.Fly specimens were totally submerged in 400 L volume of lysis buffer solution (100 mM NaCl, 10 mM Tris, 5 mM EDTA, 2% SDS, and 0.2 mg/mL proteinase K) and incubated for 4 hours at 55 ∘ C without shaking.Although overnight incubation may guarantee a better DNA yield, as it did for Gilbert et al.[12], we adopted the incubation time from Rowley et al.[13], who included thinat room temperature.Postextraction photographs were taken in the same manner as the preextraction photographs.2.4.Polymerase Chain Reaction (PCR).Primer pairs for cytochrome c oxidase subunit I (COI) gene and ribosomal internal transcribed spacer 2 (ITS2) were chosen for assessment of PCR ampli�cability from mitochondrial and nuclear DNA, respectively.Primer sequences and PCR amplicon sizes are listed in