Efficient isolation method for high‐quality genomic DNA from cicada exuviae

Abstract In recent years, animal ethics issues have led researchers to explore nondestructive methods to access materials for genetic studies. Cicada exuviae are among those materials because they are cast skins that individuals left after molt and are easily collected. In this study, we aim to identify the most efficient extraction method to obtain high quantity and quality of DNA from cicada exuviae. We compared relative DNA yield and purity of six extraction protocols, including both manual protocols and available commercial kits, extracting from four different exoskeleton parts. Furthermore, amplification and sequencing of genomic DNA were evaluated in terms of availability of sequencing sequence at the expected genomic size. Both the choice of protocol and exuvia part significantly affected DNA yield and purity. Only samples that were extracted using the PowerSoil DNA Isolation kit generated gel bands of expected size as well as successful sequencing results. The failed attempts to extract DNA using other protocols could be partially explained by a low DNA yield from cicada exuviae and partly by contamination with humic acids that exist in the soil where cicada nymphs reside before emergence, as shown by spectroscopic measurements. Genomic DNA extracted from cicada exuviae could provide valuable information for species identification, allowing the investigation of genetic diversity across consecutive broods, or spatiotemporal variation among various populations. Consequently, we hope to provide a simple method to acquire pure genomic DNA applicable for multiple research purposes.


| INTRODUCTION
Nondestructive sampling methods for DNA resources have recently attracted more attention from ethological, conservational, and population genetic studies. DNA extraction from specimens usually required scarifying essential sections of the insects such as leg, thorax, or head capsule. Such sampling methods could cause severe impacts on the species at both individual and population levels. Invasive sampling could have negative consequences on subsequent behavior and survival of sampled individuals. Extensive sampling is problematic for small colonies of social insects (Starks & Peters, 2002). Moreover, lethal sampling potentially decreases population size and alters population structure (Starks & Peters, 2002), which is harmful for the conservation of endangered species. Consequently, nondestructive sampling methods are in need for various genetic analyses (Châline, Ratnieks, Raine, Badcock, & Burke, 2004;Su et al., 2007).
One of the main reasons for the rare application of cicada exuviae in molecular works is that the exoskeleton itself does not contain any genomic material. The cuticle plays the role of the insect exoskeleton, which is chemically composed of chitin, a polysaccharide polymer of N-acetyl-glucosamine, cuticular proteins, cuticular lipids, phenols, and quinones (Nation, 2008). Trace genomic DNA can be extracted from muscle tissues or metabolic waste products that the individual left on the inner side of the exoskeleton after molt (Nation, 2008). Another reason for the rare application of cicada exuviae is the presence of potential polymerase chain reaction (PCR) inhibiting substances in soil, such as humic acids. Genomic DNA extracted from cicada exuviae can therefore contain contaminants that inhibit the usage of those DNA samples in downstream applications such as amplification of target sequence (Baar et al., 2011;Braid, Daniels, & Kitts, 2003;Kermekchiev, Kirilova, Vail, & Barnes, 2009;Schrader, Schielke, Ellerbroek, & Johne, 2012;Straub, Pepper, & Gerba, 1995).
Our goals in this study are to evaluate protocols for DNA extraction from cicada exuviae regarding their quality and quantity of DNA yield and to suggest the best protocol for downstream applications. Six extraction protocols including available commercial kits and manual protocols were tested. We further identified those parts of the cicada exoskeleton from which high DNA yield was obtained.

| Sample collection
Exuviae of the black cicada (Cryptotympana atrata, Fig. 1) were collected in Seoul,Korea (37.533415°N,127.070493°E), on 11 July 2015. The sampling location was an apartment complex where multiple cicada species coexisted, that is, C. atrata, Hyalessa fuscata, and Meimuna opalifera. After field collection, samples were identified for species based on morphological characters (Lee et al., 2012) and were stored at ambient temperature. DNA extraction work on those exuviae was performed approximately 14 months after field collection.
All samples were homogenized using a pestle. To standardize among protocols, we incubated all samples in a thermo-shaker at 2.5 xg for 20 hr. Cell lysis buffer and procedure of each protocol are shown in Table 1. For PS samples, additional 10-min vortex mixing at maximum speed using a MO BIO Vortex Adapter was performed after incubation. The remaining exoskeletons were removed from each tube after incubation, and the tubes were centrifuged at 18,000 xg for 2 min.
Each supernatant was carefully transferred to a new 1.5-ml tube, avoiding the transfer of the pellet. Subsequent steps were performed following manufactures' protocols for kits. For EtSC samples, precipitation of cell debris was performed by adding 166.7 μl of 6 mol/L NaCl to each tube followed by centrifuging at maximum speed for 10 min, after which the top supernatant layer was transferred to a new 1.5ml tube with 1 ml of cold 100% ethanol and incubated overnight at −20°C. The samples were washed twice by adding 800 μl 70% ethanol, via briefly vortexing the sample followed by carefully pipetting off the supernatant without dislodging the DNA pellet. Pellets were left F I G U R E 1 The black cicada (Cryptotympana atrata). This species is very common in urban areas in Korea. Photograph credit Yoonhyuk Bae to dry in air for approximately 30 min and then resuspended in ultrapure water (Biosesang Inc., Gyeonggi-do, Republic of Korea). For EtAA samples, a precipitation step was carried on via addition of 200 μl of 4 mol/L ammonium acetate followed by centrifugation at 18,000 xg for 20 min before transferring the top supernatant layer into a new 1.5-ml tube. The samples were washed with ethanol as described for EtSC samples and resuspended in ultrapure water (Biosesang Inc., Gyeonggi-do, Republic of Korea). Following the Ch5% protocol, the samples after incubation were further incubated at 100°C for 15 min and then centrifuged at 14,000 xg for 4 min, and the top layer supernatant was transferred to a new 1.5-ml tube. ratios ranged between 1.8 and 2.0. For Ch5% samples, due to lack of baseline buffer, we used original Chelex 5% as baseline buffer, and the measurement of two ratios was employed only for purity comparison among protocols, but was not included in the statistical analysis.

| PCR amplification and purification
Five hundred bp of the 16S region was amplified using two primers: LR-J-12887 (5′-CCGGTCTGAACTCAGATCACGT-3′) and LR-N-13398 (5′-CGCCTGTTTAACAAAAACAT-3′) (Simon et al., 1994) by Takara Ex Taq (Takara Korea Biomedical  Gyeonggi-do, Republic of Korea). PCR amplification initiated by 1-min initial denaturation at 94°C, followed by 30 cycles of 30 s denaturation at 94°C, 1-min annealing at 56°C, and 1-min elongation at 72°C, finally completed by 2-min terminal elongation at 72°C. Three microliters of each PCR product were loaded on 1.5% agarose gel and visualized using the same loading dye and LED illuminator as described above. Samples with bands that appeared at 600-bp size, as in the positive control band (Fig. 2), were considered as PCR success and were used for the gel purification procedure.
We labeled 1 for successful amplification and 0 for amplification failure. Gel bands were excised using a sterile scalpel, and gel purification was conducted using a QIAquick ® Gel Extraction Kit (QIAGEN Group, Hilden, Germany). All samples were sequenced both in forward and in reverse directions by COSMO Genetech Company (COSMO Genetech Co., Ltd., Seoul, South Korea), and sequencing success was labeled 1 as successful sequencing and 0 for sequencing failure.

| UV-Vis spectra of DNA samples
UV-Vis spectra of 40 DNA samples extracted by six protocols are shown in Fig. 3. Among those protocols, only samples extracted by PowerSoil DNA Isolation kit (Fig. 3f) show clear peaks at 260 nm, which corresponds to the absorbance wavelength of DNA, as well as humic acids available in soil.
A comparison across protocols (  Fig. 4b) showed significantly higher DNA concentrations in legs and abdomen compared to head (p < .05) and thorax (p < .05). No difference in DNA concentration was found between legs and abdomen (p > .05) or between head and thorax (p > .05).
Only samples extracted using the PS protocol possessed an A260/280 ratio within 1.8-2.0 purity range (1.89 ± 0.03, estimated mean ± SE), and they were also significantly higher than other samples in this ratio (p < .001) (  Fig. 5b); the ratio of thorax was significantly higher than that of abdomen (p = .037).

| Amplification/sequencing success
We compared all sequences to the sequence GU344091, which is a partial sequence of 16S large subunit ribosomal RNA gene of

| DISCUSSION
In this study, we compared six methods for the DNA extraction from cicada exuviae. Among those, PowerSoil DNA Isolation kit was the only extraction method that provided bands of the expected size and successful sequencing results. Although other protocols could generate high DNA quantities (Fig. 4), only DNA samples extracted with the PowerSoil kit could be amplified via PCR application (12 in a total of 40 samples). The success of PCR and sequencing did not depend on the used exoskeleton parts.
UV-Vis measurements to determine DNA concentration were performed as shown in Fig. 3. DNA concentrations were determined from the absorbance at 260 nm, which is the wavelength at which nucleic acids show an absorption maximum. Although samples extracted by ethanol precipitation methods showed a high amount of DNA according to UV-Vis measurements, such results were likely to be overestimated due to cross-absorbance of humic acids at 260 nm.
That type of contamination is commonly found in soil samples and usually coextracted with genomic DNA during the extraction procedure. Without proper separation techniques, the amount estimated by absorbance at 260 nm potentially included both genomic DNA and humic acids.   Subsequently, we expect our results will aid in genetic research of cicadas in the future.

ACKNOWLEDGMENTS
This study was financially supported by a grant from the National Research Foundation of Korea (2015R1A4A1041997).

AUTHOR CONTRIBUTION
Hoa Quynh Nguyen contributed to research conception and design, conducted statistical analysis and data interpretation, and drafted the manuscript. Ye Inn Kim prepared data acquisition. Amaël Borzée contributed to research conception and design and provided critical revisions. Yikweon Jang provided critical revisions. All authors declare to have no conflicts of interests on this manuscript.