Diversity and DNA Barcode Analysis of Chironomids (Diptera: Chironomidae) from Large Rivers in South Korea

Simple Summary We aimed to identify chironomid species collected from four large rivers in South Korea, construct a corresponding DNA barcode library, and examine the distribution and community structure of the identified riverine species. Adult chironomids were identified morphologically, and their COI nucleotide sequences were used to verify species identification and construct a DNA barcode library. The resulting COI library effectively discriminated >90% of riverine Chironomidae in South Korea. The distributional aspects of chironomid species in the four large rivers of South Korea are also discussed. Abstract Most large rivers in South Korea run through major cities, which often experience many environmental problems, including outbreaks of non-biting midges (Diptera: Chironomidae). However, chironomid species inhabiting large rivers have not been thoroughly investigated. We aimed to identify chironomid species collected from the four main large rivers in South Korea, construct a corresponding DNA barcode library, and examine the distribution and community structure of the identified riverine species. Adult chironomids were collected from nine sites along the rivers by using sweep nets and light traps during June and August 2015. Adults were morphologically identified, and COI nucleotide sequences were generated to verify the species identification and construct a DNA barcode library. The distribution and community structure of the identified species were also analyzed. A total of 124 COI sequences were established from 37 species belonging to 19 genera, and the resulting DNA barcode library effectively discriminated >90% of riverine Chironomidae in South Korea. Ten species, which are considered indicator species for large rivers, were collected from all four rivers. In addition, members of the subfamily Chironominae were collected more frequently than members of other subfamilies, with Tanytarsus tamagotoi being the most common and widespread chironomid species in South Korea. The DNA barcode library developed in this study will facilitate environmental studies of large rivers, such as biomonitoring chironomid larvae.

Chironomid larvae inhabit the most organically rich sediments of large rivers and are crucial benthic macroinvertebrates for biomonitoring rivers that supply freshwater resources to large cities or metropolises. In South Korea, these rivers include the nation's four main large rivers: the Han River (Seoul: population 9.6 million), Geum River (Daejeon: 1.5 million), Yeongsan River (Gwangju: 1.4 million), and Nakdong River (Busan: 3.3 million; Daegu: 2.3 million) [14] (Figure 1). Recent environmental changes in river systems, such as increases in lentic areas due to dam construction, could result in increasingly favorable environments for chironomid larvae that prefer organic-rich fine sediments that are often present in areas with slow currents and in the lentic areas of large rivers [15][16][17]. For example, 16 massive concrete weirs were recently constructed as part of the "Four Major Rivers Project" (construction period 2009-2011) in South Korea, and after this construction, >90% of the benthic macroinvertebrate fauna in the corresponding impoundment areas were replaced, in terms of richness and abundance, by chironomid larvae and oligochaete worms [18]. Additionally, larval chironomids are sometimes found in water purification plants and tap water in Korea [19].
With an increase in chironomid larvae, the emergence rate of adults also increases, causing a nuisance to people. Although adult chironomids do not bite or spread diseases as mosquitoes do, they occur in large numbers in rivers and streams. Public health is negatively affected by debris in the form of chironomid remains in these water bodies that can contribute to respiratory allergies and the breeding and transport of pathogens [20].
Despite the ecological and environmental importance of riverine chironomids, studies on chironomids in Korea have been limited to the taxonomy of specific groups [6,10,21,22] or local fauna [9,11], and intensive investigations of riverine chironomid species have yet to be conducted. To facilitate such investigations, fully evaluating the diversity of riverine chironomids and accurately identifying chironomid larvae during large river biomonitoring in South Korea are necessary, and a chironomid DNA barcode library, based on adult chironomid specimens from South Korea's main large rivers, is critical.
In this study, we aimed to identify chironomid species collected from the four main large rivers in South Korea, construct a corresponding DNA barcode library, and examine the species distributions and community structures of the identified riverine species. We expect the resulting DNA barcode library to be a useful reverse taxonomy tool for those involved in diverse taxonomic and ecological studies of chironomids in the future.

Sampling Locations and Methods
South Korea's four main large rivers, namely the Han River (length: 481 km; basin area: 26,018 km 2 ), Geum River (length 397 km, basin area 9810 km 2 ), Yeongsan River (length 116 km, basin area 2798 km 2 ), and Nakdong River (length 521 km, basin area 23,871 km 2 ) [23], provide the most freshwater resources to South Korea's major cities ( Figure 1). Chironomid sampling was performed at nine sites along these four major rivers (Figure 2a-d), and all sampling sites were located between 0.5 and 1 km upstream of newly constructed weirs to ensure that adult chironomids collected at the sampling sites accurately represented the chironomid fauna of the rivers' lentic habitats.
In June and August 2015, adult chironomids were sampled using a 30 cm diameter sweep net (10 sweeps from grass along a 10 m transect; Figure 2e) for quantitative analysis and using a light trap (Figure 2f) as well as a sweep net for qualitative purposes, and then they were preserved in 80% or 99.5% ethanol.

Morphological Identification
Some chironomid genera are difficult to identify at the species level without slide mounts of specific body parts; therefore, the antennae, head, wings, abdomen, and hypopygium of the adult samples were dissected using a fine needle under a dissecting microscope (SZ61; Olympus, Tokyo, Japan). To prepare permanent slides, the body parts were mounted with Hoyer's medium and dried for 2-4 days [24]. All specimens were identified using available identification keys and references [5][6][7]. The remaining body parts of each adult specimen were preserved in 99.5% ethanol prior to DNA extraction. All specimens were deposited at the Entomological Museum of Korea University (KU) in Seoul, Korea. The terminology used in this study generally follows that of Saether [25], Oliver and Dillon [26], and Langton and Pinder [27].

Sequence Generation and Analysis
To verify the correctness of the species identification of the morphologically classified midges and to construct DNA barcodes, total genomic DNA was extracted from the thoracic pre-episternum of each specimen using a DNeasy Blood and Tissue kit (Qiagen, Hilden, Germany). Standard PCR amplification and sequencing protocols were used to generate COI fragment sequences. Briefly, the target fragment of COI was amplified in 20 µL reactions containing AccuPower PCR PreMix (Bioneer Co., Daejeon, Korea), 1 U Top DNA polymerase, dNTPs (10 mM), Tris-HCl (pH 9.0), KCl (30 mM), MgCl2 (1.5 mM), 1-3 µL (5-50 ng) template DNA, and 1 µL of each primer (LCO1490 and HCO2198; 10 pM each) [28]. Amplification was performed using the following thermal cycling program: 94 • C for 5 min; 35 cycles of 94 • C for 0.5 min; 48 • C for 1 min; 72 • C for 1.5 min; and a final extension at 72 • C for 10 min. The reaction products were separated on 1.5% agarose gels, visualized using UV light, purified, and then sequenced using an ABI Prism 3130 Genetic Analyzer (Applied Biosystems, Foster, CA, USA). The sequences were aligned using the CLC Main Workbench (version 7.8.1; CLC bio, Aarhus, Denmark), and the results were crosschecked using the ClustalW algorithm in MEGA 7.0 [29]. For phylogenetic relationships between large Korean river chironomid species, a maximum likelihood (ML) tree was constructed using MEGA 7.0, and the GTR + I + Γ model of nucleotide substitution was selected as the best-fitting model for the COI sequences using jModelTest (version 2.1.10) based on the Akaike information criterion [30,31]. The convergence diagnostic was calculated for 1000 generations in ML, and groups with a frequency greater than 50% were retained.
Pairwise sequence divergence within and between genetic clusters was calculated using the K2P model in MEGA 7.0. Intraspecific and interspecific genetic distance analyses were performed using the K2P and maximum composite likelihood models of Tamura et al. [32], which were the best-fit nucleotide substitution and base frequency models, respectively, as indicated by MEGA 7.0.
A model-based species depiction analysis showed that species identification did not depict species. However, the phylogenetic tree formed a mono-clade, and the average barcode gap of native chironomids was 3% [33]. In addition, for Tanytarsus, which has many unknown and cryptic species, Lin et al. [34] proposed a threshold of 4-5% for the species. Therefore, we identified species that were higher than commonly known thresholds. Therefore, even if model-based species depictions have not been tested, there is little chance of species classification error.
Sequences of COI of Korean chironomids from large rivers obtained in the current study were deposited in GenBank under accession numbers OM974370-OM974448.

Chironomid Community Analysis
To evaluate the structure of the chironomid community in the study area, the dominance index (DI), diversity index (H'), richness index (RI), and evenness index (J'), which indicate the quality of the community structure, were calculated as the proportion of the relative density of the primary and secondary species in the community [35] based on the Shannon-Weaver function [36] and by using the formula of Margalef [37] and Pielou [38], respectively.

Morphology-and DNA-Based Species Identification
A total of 2168 individuals belonging to 40 species, 21 genera, and 3 subfamilies of Chironomidae were collected and identified by quantitative sampling. The identity of most species distinguished morphologically (n = 37) was confirmed using COI sequence analysis. However, based on morphological and molecular data, one species (Chironominae) could not be assigned as a valid chironomid species. Only three of the remaining species (Cladopelma edwardsi, Corynoneura sp., and Einfeldia pagana) could be identified by morphology alone, owing to a lack of sufficient specimens for DNA analysis. Furthermore, most of the species (n = 30) belonged to genera of the subfamily Chironominae, namely, Benthalia (one species), Chironomus (four species), Cladopelma (one species), Cryptochironomus (one species), Dicrotendipes (three species), Einfeldia (one species), Endochironomus (one species), Glyptotendipes (one species), Harnischia (one species), Lipiniella (one species), Microchironomus (one species), Parachironomus (two species), Phaenopsectra (one species), Polypedilum (seven species), Tanytarsus (three species), and an unknown genus (one species). Meanwhile, six species belonged to the genera of the subfamily Orthocladiinae, namely Corynoneura (one species), Cricotopus (four species), and Nanocladius (one species), and four species belonged to the genera of subfamily Tanypodinae, namely Ablabesmyia (two species), Procladius (one species), and Tanypus (one species). Three of the identified species (Polypedilum kyotoensis, P. okiharaki, and P. sordens) were new additions to the documented chironomid fauna of South Korea (Table 1). No. of species found in each river 25 28 18 29 M (morphology), D (DNA barcode), U (unknown), + (species found in the indicated river), and N (the number of species; the reference sequences are in parentheses). The names of the rivers correspond to the abbreviations used in Figure 1. Asterisks indicate species recorded as new to the Korean fauna. GenBank accession numbers with daggers ( † ) are specimens with sequences acquired from GenBank.

DNA Barcode Library
A total of 124 COI sequences from 37 species were generated from 81 collected specimens (Han River = 23, Geum River = 26, Yeongsan River = 14, and Nakdong River = 18) and combined with 43 sequences from the GenBank database to establish a DNA barcode dataset (Table 1). An analysis of the sequences confirmed the monophyly of all the sampled subfamilies, except Orthocladiinae, since the analysis indicated that the Orthocladiinae species Nanocladius tamabicolor belonged to Tanypodinae. In addition, clustering of the 81 COI sequences in the ML tree was congruent with the accepted delineations for most morphospecies. Clades corresponding to 33 (91.7%) of the 36 chironomid species (excluding one invalid Chironominae species) were strongly supported, with a bootstrap value of >95% (Figure 3). Exceptions occurred for several taxa in which the level of deep sequence divergence was observed between individuals assigned to the same morphospecies (Benthalia carbonaria, Dicrotendipes nervosus, and Ablabesmyia longistyla; Figure 3).
The range of intraspecific divergence values for the studied species was 0-15.0%, with an overall mean of 3.0% (SE = 0.6%). Minimum intraspecific divergence (0%) was observed in Polypedilum japonicum and Tanypus punctipennis, whereas high levels of intraspecific divergence (>2.3%) were observed in 15 of the 36 species, with the maximum intraspecific divergence (15%) observed in A. longistyla. The minimum interspecific divergence (0.5%) was observed between Cricotopus tricinctus and Cricotopus sylvestris, and the maximum interspecific divergence (28.4%) was observed between Polypedilum nubifer and Chironominae species. These include species complexes for which the taxonomic status was unresolved by morphological methods.
The mean and standard deviation of the chironomid community indices are listed in Table 2. The DI of the Geum River (0.78 ± 0.20), which accounted for more than 40% of the total population, was the highest among the four rivers, whereas that of the Nakdong River (0.51 ± 0.12) was the lowest. However, H', which indicates relative community balance and complexity, was the highest in the Nakdong River community (2.70 ± 0.46) and the lowest in the Geum River community (1.76 ± 0.77). Similarly, both RI and J' were lower in the Geum River community (1.43 ± 0.84 and 0.78 ± 0.16, respectively) than in the other river communities.
The most dominant chironomid species throughout the nine sampling sites was T. tamagotoi, while L. moderata, M. tener, Tanytarsus formosanus, and P. choreus were locally dominant at some of the sampling sites (Table 2).

Discussion
Freshwater biologists commonly use chironomid larvae to assess and monitor environmental conditions, particularly the levels of organic pollution in streams, rivers, and wetlands [39][40][41]. Unfortunately, chironomid larvae are usually difficult to identify morphologically, and species identification mostly relies on an analysis of adult males, which tend to possess more species-specific characteristics [1,16]. Accordingly, the DNA barcode library developed in the present study may facilitate the accurate identification of larval chironomids. Moreover, even though COI sequences for some Korean chironomids have been fragmented in previous studies [33], the DNA barcode library developed here for chironomids of large South Korean rivers could facilitate a wide variety of biological and environmental projects, such as biomonitoring of large rivers. In addition, 31 species of chironomids were morphologically classified in this study; however, the usefulness of the DNA barcode library was clearly demonstrated by identifying six additional species for a total of 37 species.
In this study, 40 chironomid species, comprising~13% of the chironomid species distributed in South Korea, were identified using morphological and molecular methods. The composition of predominant chironomid groups in rivers and lowland streams is well known, comprising the subfamilies Chironominae, Orthocladiinae, and Tanypodinae [9,39]. Chironominae was the predominant group in this study, followed by the subfamilies Orthocladiinae and Tanypodinae. However, Podonominae and Diamesinae subfamilies, as well as certain groups of Orthocladiinae, were not included in this study because we focused on midges inhabiting rivers located in urban centers and did not include headwaters or pristine waterways [42].
Among the 40 species collected, the species of the genera Chironomus (C. circumdatus, C. kiiensis, C. nipponensis, and C. plumosus), Glyptotendipes (G. tokunagai), Polypedilum (P. nubifer), Tanytarsus (T. formosanus, T. oyamai, and T. tamagotoi), Cricotopus (C. sylvestris), and Procladius (P. choreus) were collected in large numbers using a light trap, which may indicate that these species are responsible for nuisance activity, such as disturbances in city lights and of outdoor activities. In addition, the genera Chironomus, Polypedilum, Tanytarsus, and Cricotopus occur more frequently in polluted rivers in Asia [39], as observed in this study. Furthermore, Tanytarsus is the most abundant species across the four major river sites and has been used as an indicator of moderately polluted streams [43]. In this study, the three Polypedilum species (Chironominae) recorded for the first time in Korea were identified by DNA barcoding.
The average intraspecific divergences (0.9-2.32%) were previously reported in Chironomidae [44,45]; however, some researchers have suggested a cryptic diversity range of 2-5% [46,47] or an average intraspecific threshold of 4-8% for members of the genera Tanytarsus and Polypedilum [34,48]. In this study, more than 8% intraspecific divergence was observed in 3 (D. nervosus, M. tener, and A. longistyla) of the 37 species. Cases of deep intraspecific divergence can reflect misidentifications, cryptic taxa, ancestral polymorphisms, or introgression [49]. No significant morphological differences were found between the species divided into the two groups of D. nervosus (Figure 3). Therefore, this species includes cryptic species. Ablabesmyia longistyla also exhibited 15% divergence at the divergence level of the Korean species. Ablabesmyia includes several other cryptic species, and the final resolution requires a detailed taxonomic study of the entire group. By contrast, C. sylvestris and C. tricinctus were grouped together ( Figure 3). Considering that the average genetic distance between C. sylvestris and C. tricinctus was 0.5% (see File S1), these specimens could be the same species or two closely related sibling species. From a taxonomic perspective, Cricotopus is one of the most difficult genus in Chironomidae to identify because of its lack of reliable diagnostic characters [45]. In general, COI barcodes for each species formed a distinct cluster separated from its nearest neighbor, but there were exceptions. Some of these cases involved unusually large intraspecific distances, whereas in other cases, there was little or no separation between species. Where barcodes failed to distinguish between species, the taxa involved were generally morphologically similar and closely related [49]. Therefore, species with a limit of COI barcode analysis such as those of Cricotopus require specialist entomologists' support and the collection of as many representatives of each type as possible to obtain reasonable estimates of intraspecific variation [50,51].

Conclusions
A total of 124 COI sequences were established from 37 species belonging to 19 genera, and the resulting DNA barcode library effectively discriminated >90% of riverine Chironomidae in South Korea. Ten species, which are considered indicator species for large rivers, were collected from all four large rivers, and members of Chironominae occurred more frequently than members of other subfamilies, with Tanytarsus tamagotoi being the most common and widespread species in large rivers in South Korea. This study contributes to the current knowledge of riverine chironomids, which have the potential to cause environmental and public health problems for residents of large cities in Asia, including South Korea.