The genus Osmoderma LePeletier de Saint-Fargeau and Serville consists of 12 species, which occur throughout the Palearctic (9 species) and Nearctic (3 species) regions (Audisio et al., 2007). Species of Osmoderma are commonly referred to as “hermit beetles” because they live in hollows in old broadleaved trees and are rarely observed in nature (Audisio et al., 2009). European species of these beetles (mainly O. eremita) are regarded as one of the most important “flagship species” and “umbrella species” for the conservation of local saproxylic communities (Ranius, 2002a, 2002b; Audisio et al., 2009). Among these, five species are listed as near threatened to endangered in the IUCN Red List of Threatened Species by virtue of their rarity and association with a specific habitat type (Nardi and Micó, 2010; Nieto and Alexander, 2010).
In the Korean Peninsula, there are a total of four species and one subspecies in the genus Osmoderma. O. opicumLewis, 1887 has been recorded in South Korea by Cho (1969), whereas three species, O. barnabitaMotschulsky, 1845 described by Stebnicka in 1980, O. davidis forma amurensisGusakov, 2002 described by Gusakov (2002), and O. caelesteGusakov, 2002 described by Audisio in 2007, are listed in North Korea, although without any reliable specimen identification. Furthermore, recently, a new subspecies, O. opicum coreanum, was also described from North Korea by Tauzin (2013).
It is noteworthy that South Korean specimens of Osmoderma have been extremely rarely observed. Only four specimens, consisting of three females and a male, have been collected in South Korea since 1969 (Kim, 2011). The South Korean population, which is considered to be O. opicum, has accordingly been registered as a critically endangered species in the Korean Red Data Book (Lee, 2013). In order to determine accurate species traits of the South Korean hermit beetle, in 2012, we first obtained a Korean female specimen, which we assumed to be O. opicum (specimen no. 3257) and six authenticated Japanese specimens of O. opicum. However, these two geographically separated populations differed in several of their external features. It was accordingly suggested that the South Korean hermit beetle might have been incorrectly recorded as O. opicum. Therefore, over the past 4 years, we have attempted to obtain a larger number of specimens of the Korean hermit beetle as well as voucher specimens of the close relatives in neighboring regions. Recently, three appreciable studies on the European hermit beetles, carried out mainly using molecular and distributional data, have contributed to a reasonable and objectively supported taxonomic arrangement (Audisio et al., 2007, 2009; Landvik et al., 2013). It was therefore possible to compare our molecular data with the results obtained for European hermit beetles.
The South Korean hermit beetle is an important insect species in the conservation management of threatened species comprising forest saproxylic communities. Nonetheless, as mentioned above, there is still a lack of accurate information on the exact species identity and ecological traits of the Korean population that we seek to conserve.
Materials and Methods
Specimen collection and morphospecies examination
To assess the exact species status of the South Korean hermit beetle, we obtained a total of 16 dried specimens belonging to three Osmoderma species: six South Korean specimens, which have been recognized as O. opicum, two of which were previously recorded as voucher specimens (nos. 7299 and 9572) by Kim (2011); five specimens of true O. opicum from Honshu, Japan; three specimens of true O. caeleste from Far East Russia; and two specimens of O. barnabita from West Russia. We were, however, unable to obtain any specimens of O. opicum coreanum and O. davidis (Table 1).
For simultaneous examination of both morphological and molecular characters, we first observed the general structure of the specimens under a stereoscopic microscope (MZ 16A and MZ 6; Leica, Solms, Germany) for morphospecies identification and diagnostic characteristic delimitation. All of the examined materials and genomic DNA stocks are preserved in the insect collection at the National Institute of Agricultural Science, Jeonju, Korea. The provincial abbreviations of the collecting sites within Korea are as follows: GW, Gangwondo; GB, Gyeongsangbuk-do.
DNA extraction, PCR strategy, and DNA analyses
To reconfirm the specific status of the specimens, we performed molecular analysis using sequences of the cytochrome c oxidase subunit I (COI) gene, which is useful in identifying species within a closely related species group (Han et al., 2012). For genomic DNA extraction, we removed the genitalia from the abdomens of both sexes and used them directly to extract genomic DNA using non-destructive methods (Han et al., 2012). The remaining genital structures were cleaned by heating with 10% KOH solution in a Wise Therm ®HB-48P heating block at 60℃ for 1-2 h and then preserved in microvials containing glycerine for morphological examination.
For PCR based on dried specimens, we first attempted to amplify the DNA barcoding region (658 bp) of the COI gene using the universal primer set LCO1490/HCO2198 (Folmer et al., 1994); however, we were unable to obtain amplicons for most samples; this should be expected considering the likelihood of DNA degradation in specimen maintained under unknown dried conditions (Goldstein and Desalle, 2003; Hajibabaei et al., 2006; Wandeler et al., 2007). To retrieve the barcoding regions of the COI gene, we adopted the PCR methodology focused on old museum specimens, as described by Han et al. (2014), and designed 13 new specific primers for the COI gene of O. opicum and O. caeleste using Primer3 (Untergrasser et al., 2012) (Table 2). PCR was conducted using AccuPower ProFi Taq PCR PreMix (Bioneer, Daejeon, Korea). The amplification profile consisted of pre-denaturation for 5 min at 95℃, 35-45 cycles of 30 s at 95℃ for denaturation, 30 s at 48-53℃ for primer annealing, and then 45 s at 72℃ for extension, with a final extension for 5 min at 72℃. To confirm product size and purity, the PCR amplicons were assessed by 0.7% agarose gel electrophoresis using Safe-Pinky (Gen DEPOT, Barker, TX, USA). The amplicons successfully obtained were purified using the AccuPrep PCR Purification Kit (Bioneer, Daejeon, Korea) following the manufacturers protocol. Sequencing was performed commercially by Macrogen (Seoul, Korea). All sequences were generated In both directions. The quality and potential polymorphic sites of the analyzed sequences were checked using Chromas 2.33 (Technelysium Pty Ltd, Australia).
COI sequence analysis
In this study, sixteen COI sequences from 16 dried specimens were successfully generated. In addition, 32 COI sequences of five species published in two previous studies (Audisio et al., 2009; Landvik et al., 2013) were downloaded from the GenBank (http://www.ncbi.nlm.nih.gov/genbank) (Table 1). Data sets of the nucleotides of COI genes were aligned in MEGA 5.2 (Tamura et al., 2011) using ClustalW with the default settings (Gap Opening Penalty = 15, Gap Extension Penalty = 6.66 in both pairwise and multiple alignments). The anterior and posterior regions of uncertain alignment were eliminated from the data matrix. The COI alignment for reading frames was checked manually by translating sequences into amino acids to identify stop codons and potential shifts. Sequences were finally trimmed to 657-bp fragments. COI sequences generated in this study are available from GenBank under the accession numbers KX346561-76 (Table 1). For molecular species identification, a neighbor-joining (NJ) analysis (Saitou and Nei 1987) was performed using MEGA 5.2. Genetic distances within and between species of hermit beetle were calculated using Kimura’s two-parameter test (Kimura, 1980) in accordance with Nei’s empirical guidelines (Nei 1991, 1996).
Results and Discussion
Morphospecies identification of the Korean hermit beetle
We examined diagnostic morphological features for each specimen and species. The South Korean hermit beetle (Fig. 1 A and B) was revealed to be identical to the true O. caeleste from Far East Russia (Fig. 1C and D). These taxa share the same characteristics, such as an arched frontal margin of the head, a reverse triangular depression on the pronotum (Fig. 2A-D) and shortly developed parameres of the aedeagus (Fig. 3A and B). In contrast, O. caeleste can easily be distinguished from the true O. opicum on the basis of the truncated frontal margin of the head, the longitudinal and narrow groove on the pronotum (Fig. 2E and F), and the distinctly elongated parameres of the aedeagus (Fig. 3C). O. barnabita (Fig. 1G and H) is also distinctly separated from the South Korean hermit beetle by the darker chestnut color of the body, the longitudinal and narrow groove, the rather spare punctures on the pronotum (Fig. 2G and H), and the shortly developed, but with the lobe likely expanded base of parameres of the aedeagus (Fig. 3D). In the Far East Asia, additional related species, O. davidis is distributed in Russia, E. China and Mongolia. The South Korean hermit beetle was compared with this species according to the literatures by Boucher (2002) and Tauzin (2013). O. davidis has the ginkgo leaf-liked pronotal groove and the elongate parameres of the aedeagus. It is indicated that the South Korean hermit beetle and O. davidis are a distinctly separated species, respectively.
Tauzin (2013) described a subspecies, Osmoderma opicum coreanum, from Hamgyong province, North Korea based on a male holotype, which was provided Dr. Jingke Li (see Tauzin 2013, p. 2). However, we must point out to have doubt on the original locality of the specimen of the new subspecies, because Dr. Li has often provided to purchasers many insect specimens collected from adjacent countries of Asia such as China, Russia, and Japan as though they were from North or South Korea. The existence of this subspecies in Korea is needed to reconfirm in the further study.
COI sequence analysis
A total of 16 COI sequences representing three morphospeices, O. opicum, O. barnabita, and O. caeleste including the Korean population, were successfully generated. Before analyzing the COI sequences, we confirmed the absence of any putative Numts and heteroplasmies from the analyzed sequences. One sequence (specimen no. 7299), only partly sequenced in 226 bp fragment, was confirmed to be identical with the other sequences of the Korean hermit beetle, but not including in the calculation of genetic distance and in NJ analysis.
The NJ tree topology (Fig. 4) and the pairwise distances of COI sequences (Table 3) showed that the Korean specimens were strongly clustered to those of the Far East Russian O. caeleste with low intraspecific (range: 0-0.3%). But, this clade was distinctly separated from other species clades by large interspecific distances (range: 13.7-19.4%). This result indicates that the South Korean hermit beetle and the true O. caeleste from the Far East Russia are the same species. The Japanese species, O. opicum, was also solely clustered with low intraspecific (range: 0-0.8%) and large interspecific distances (range: 12.1-16.0%). The interspecific distances between O. caeleste and O. opium were significantly large, ranging from 16.6% to 17.2%. This result insinuates that these two species have undergone an independent evolution, respectively, and more early separated from the extant European species. Two West Russian specimens of O. barnabita analyzed in this study were clustered with the previous sequences of O. barnabita (Audisio et al., 2009; Landvik et al., 2013). Therefore, the South Korean hermit beetle is identified to O. caeleste based on both of morphological and DNA barcoding analyses.
Systematic Accounts
Family Scarabaeidae Latreille
Subfamily Cetoniinae Leach
Tribe Trichiini Fleming
Genus Osmoderma LePeletier and Audinet-Serville
Osmoderma caeleste (Gusakov, 2002)
Osmoderma barnabitum var. castaneumTauzin, 1994: 204; Tauzin, 2006: 43 and 44.
Gymnodus caelestisGusakov, 2002: 13 and 39 (Primorski Krai, Far East Russia).
Osmoderma sikhotenseBoucher, 2002: 426 (Sikhote-Aline, Far East Russia).
[Korean records]
Osmoderma opicum Lewis: Cho, 1969: 688 (Atlas with a female photo, misidentification); Kim, 1978: 379 (Distributional map, listed); Shin and Yoon, 1994: 153 (Check list); Kim, 1998: 358 (List); Kim, 2001: 132 (Examined five specimens, misidentification); Kim, 2011: 33 (Examined five specimens, misidentification).
Material examined. South Korea: 1 male, Gangwon-do, Yanggu, 26. IX. 2011, collector unknown (voucher no. 3257); 1 female, Gangwon-do, Injae, Jogyeong-dong, Garigol, 26. VIII. 2000, H. N. Seok (voucher no. 7299); 1 female, Gangwondo, Injae, Buk-myeon, Yongdaeri, 31. VII. 2013, Lim et al. (voucher no. 9571); 1 male, Gangwon-do, Hwacheon, Sangseomyeon, Guunri, 9. VIII. 2000, J. R. Cha (voucher no. 9572); 1 female, Gangwon-do, Pyeongchang, Jinbu-myeon, Mt. Odae, 8. VII. 2010, collector unknown (voucher no. 7387); 1 female, Gyeongsangbuk-do, Yongyang, Ilweol-myeon, Mt. Ilweol, XI. 2008, Y. Y. Baik (voucher no. 9596).
Distribution. South Korea, North Korea (?), E. China (?), Mongolia (?), and the Far East Russia.
Remarks. We examined the female photo represented in Cho (1969) and two materials (nos. 7299 and 9572) used in Kim (2001, 2011). Based on our morphological and molecular examinations, three previous South Korean records were misidentifications of O. caeleste. Tauzin (2013) mentioned that O. caeleste is distributed in the Far East Russia, Beijing in China, Inner Mongolia and Korea. Among them, the distributions of China, Mongolia and Korea were listed having no the basis of examined specimen records. Therefore, it is needed that the exact distributional range of this species has to reexamine in the further study.