A total of 99 Apis mellifera colonies were sampled from 72 apiaries in 5 provinces (Gyeonggi, Gangwon, Chungcheong, Gyeongsang, and Jeolla) in Korea (Fig. 1). Fifty to 100 adult bees were collected and pooled from 1-2 colonies in each apiary between August and December 2013. The honey bee samples were collected from brood nests inside the hives for managed colonies and at the hive entrances for feral colonies, respectively [5]. After collecting the bees, the samples were stored at -20˚C for PCR detection.

The head and thorax of 20 adult bees randomly chosen from each hive were used for DNA extraction using DNeasy Blood & Tissue kit (Qiagen, Hilden, Germany) following the manufacturer’s instructions. Total DNA was eluted in 30 µl of elution buffer and used for PCR with LeGene Hot Start PCR Mix w/Sky Blue Dye (LeGene Biosciences, San Diego, California, USA). Two primer sets, including Acarapis (A) primer and Kojima (K) primer set, were used to identify Acarapis mite (Table 1) which were capable of amplifying mitochondrial cytochrome c oxidase subunit I (COI) DNA fragments of Acarapis mites [5,9,10]. These primers do not amplify honey bee COI DNA fragments; thus, total DNA isolated from bees was used as a template for PCR. As a positive control for PCR to verify the quality of DNA extraction, a honey bee genomic DNA fragment encoding a part of a honey bee Hymenoptera-specific transient receptor potential A channel (AmHsTRPA) was amplified by AmHsTRPA (T) primer set [10]. The thermal cycling conditions were as follows: 1 cycle of initial denaturation at 94˚C for 2 min, 35 cycles of denaturation at 94˚C for 30 sec, annealing at 55˚C for 30 sec, and extension at 72˚C for 30 sec. The PCR product was analyzed by 2% agarose gel electrophoresis. A negative control lacking template DNA was performed for each PCR reaction. Since A and K primer sets equally amplify COI genes of 3 species of Acarapis mites (A. woodi, A. dorsalis, and A. externus), the amplified products with T and A primer sets or T and K primer sets were sequenced and aligned with the sequences deposited in GenBank to determine the parasitized mite species. The nucleotide sequences were identified by the Basic Local Alignment Search Tool (BLAST) at the National Center for Biotechnology Information (NCBI). Multiple nucleotide alignment was carried out using the published SBV sequences as references by BioEdit version 7.0.9.0 (Table 2). The phylogenetic tree was constructed by Mega 6.06.1 software [11] using the neighbor-joining (NJ) method [6].

To confirm the presence of Acarapis mites, a total of 20 bees per each colony were examined with a modified dissection method developed by Lorenzen and Gary [12]. The external surface, including the head, thorax and a groove between the mesoscutum and mesoscutellum of each bee, was examined to find out A. externus and A. dorsalis. To examine A. woodi, each bee was pinned through the thorax into bee wax culture dish and placed under a dissecting microscope. The head and the first pair of legs were removed off using a scalpel or razor blade. The ring of prothoracic sclerite (collar) was also removed using fine forceps to fully expose both left and right sides of the trachea. The exposed trachea of 2 right and left sides was examined with 20× magnification to find out tracheal color change, the body of A. woodi and foreign substance. If the presence of A. woodi or foreign substance was confirmed, the trachea was detached and placed on a glass slide and examined under an optical microscope to diagnosis A. woodi.

A total of 99 colonies from 72 apiaries in 5 provinces (Gyeonggi, Gangwon, Chungcheong, Gyeongsang, and Jeolla) were examined (Fig. 1). All of the 99 colonies were T primer set positive. Among these, 36, 2, and 4 colonies were positive for A, K, and both A and K primer sets, respectively (Table 2). The infection rate of Acarapis mites in Gangwon, Gyeonggi, Chungcheong, Gyeongsang, and Jeolla were 25.0%, 9.5%, 25.0%, 95.5%, and 50.0%, respectively (Table 2). Gyeongsang had the highest infection rate, followed by Jeolla, Chungcheong, and Gyeonggi. Gangwon had the lowest infection rate among the 5 provinces. Between the primer sets, the A primer set was more sensitive than the K primer set (Table 2).

To confirm the above PCR assay results, all 42 Acarapis-positive colonies were sequenced and aligned. Each sequence showed 92.6-99.3% homology with reference sequences available in GenBank (Table 3). Species identification of each colony was determined if the sequence between the unknown sample and the reference showed 98-100% homology. Based on these criteria, the number of colonies infected with A. dorsalis was 32 samples which showed the highest infection rate among the 3 species, while the number of colonies infected with A. externus and A. woodi was 9 and 1, respectively. The DNA sequences of PCR-amplified CO1 fragments infested by A. dorsalis A. externus, and A. woodi were deposited to a DDBJ database as the no. LC006085, LC006083, and LC006084, respectively. Although A. externus and A. dorsalis were distributed throughout all 5 provinces, A. woodi was only detected from an apiary in Mungyung City, Gyeongsang (Gyeongsangbuk-do Province) (Table 4). The phylogenetic analysis between Korean isolates in this study and other references was performed (Fig. 2). Each Korean isolate Acarapis species was placed on a separate branch and closely related to the same Acarapis species.

In microscopic examinations, a total of 20 bees per each colony were dissected, but no diagnosable mite body was detected. However, foreign bodies which assumed the mite and some debris were discovered from the A. woodi-positive bee colony (Fig. 3).