DNA barcoding and morphology reveal a new cryptic species of Nagiella (Lepidoptera, Crambidae, Spilomelinae) from Japan

Abstract Nagiella tristalis Matsui & Naka, sp. nov. is described from Japan, based on DNA barcoding and morphological evidence. The two species previously known from Japan, N. quadrimaculalis and N. inferior, are diagnosed. Photographs of adults, including male and female genitalia of the three species, are provided.


Introduction
Nagiella Munroe, 1976 was established as a replacement name for Nagia Walker, 1866 (type species: Nagia desmialis Walker, 1866), which is a junior homonym of Nagia Walker, 1858 (Lepidoptera, Noctuidae). Munroe (1976) included Scopula quadrimaculalis Kollar & Redtenbacher, 1844 and Sylepta inferior Hampson, 1899 in this genus, and he described N. hortulatoides Munroe, 1976. Munroe also treated the type species, N. desmialis, as a synonym of N. quadrimaculalis. Ullah et al. (2017) studied this genus in China and described N. occultalis Ullah & Yang, 2017, which is a cryptic species of the N. quadrimaculalis species complex. Lu and Du (2020) described another species, N. bispina Lu & Du, 2020 from China. So far, this genus comprises five species, i.e., N. quadrimaculalis, N. inferior, N. hortulatoides, N. occultalis, and N. bispina. These species are often confused due to their similar appearance; in these species, the ground color is uniformly greyish and each wing has a conspicuous white spot on each wing, except for N. hortulatoides.
Two species, N. quadrimaculalis and N. inferior, have been recorded in Japan under the genera Sylepta Hübner, 1823 (Shibuya 1928(Shibuya , 1929, Nagia (Mutuura 1957), and Pleuroptya Meyrick, 1890 (Inoue 1982;Sasaki and Yamanaka 2013). In this paper, we describe and illustrate N. tristalis sp. nov., a cryptic species of the N. quadrimaculalis species complex, from Tottori Prefecture, Japan. We also provide a phylogenetic hypothesis of relationships based on the mitochondrial COI region for the three Japanese Nagiella species and N. occultalis.

Sampling insect specimens
Most specimens of N. tristalis were obtained by collecting the larvae in rolled leaves of Rubus buergeri Miq. (Rosaceae) during the winter and then rearing them by the method as described below. We also collected the adults of Nagiella species and Patania ruralis (Scopoli) (to use as the outgroup in the phylogenetic analysis) from various localities of Japan, by light-trap and daytime search. In addition, several specimens of Nagiella species were obtained by rearing eggs with the method described below.
Female moths were placed in plastic cups (Clean Cup 129 Pi 860B, with lid Clean Cup 129 Pi FSL [Risupack, Gifu, Japan]; diameter 129 mm, height 130 mm) with fresh leaves of Rubus buergeri or R. trifidus Thunb. for egg laying. The hatched larvae were reared using fresh leaves of R. buergeri or R. trifidus under a 14L:10D photoperiod at 25 ± 2 °C and 50-60% relative humidity until pupation, and the resulting pupae were kept in the same conditions until the emergence of adults.
The holotype of the new species is deposited in the National Museum of Nature and Science (NSMT; Tsukuba, Ibaraki, Japan), and the paratypes are stored in the authors' private collections.

Genitalia preparation
Before examining the male and female genitalia, the abdomen was detached from the specimen and soaked in a 10% potassium hydroxide (KOH) solution. The soaked abdomen was kept at room temperature overnight and then incubated at 60 °C for 3-6 h. After incubation, the abdomen was transferred into a glass dish with 70% ethanol, and the genitalia were detached from the abdomen under a stereomicroscope (LW-820T; Wraymer Inc., Osaka, Japan) using scissors and tweezers. The genitalia were stained with merbromin in 70% ethanol and mounted on a glass slides in Euparal. The photographs of the whole genitalia were captured with a stereomicroscope (SZX10; Olympus Corp., Tokyo, Japan) and a digital camera (DP25; Olympus Corp., Tokyo, Japan). The magnified views of genital structures were captured by an upright microscope (BX53; Olympus Corp., Tokyo, Japan) with a digital camera (DP21; Olympus Corp., Tokyo, Japan). Genital structures were measured on the screen by Fiji (Schindelin et al. 2012), based on the photographs and whole genitalia lengths measured by a ruler. As references for the terminology, we followed Kristensen (2003) and Kirti and Gill (2007) for the genitalia and Ullah et al. (2017) for the wing maculation.

DNA extraction, PCR amplification, and sequencing
Total DNA was extracted from the middle legs of the moths using the DNeasy Tissue Kit (Qiagen, Hilden, Germany). The legs were crushed using BioMasher II (FUJI-FILM Wako Pure Chemical Co., Osaka, Japan) and incubated with Proteinase K (Takara Bio Corp., Shiga, Japan) for 3-7 d at 60 °C to elute DNA. Subsequent procedures followed the manufacturer's protocol of the DNeasy Tissue Kit.
The PCR products were checked by electrophoresis on a 1% agarose gel and were purified using NucleoSpin Gel and PCR Clean-up (Takara Bio Corp., Shiga, Japan). Sequencing was conducted at Premix2 analysis service (Fasmac Co., Ltd, Kanagawa, Japan) using the primers LCO1490 (forward: GGTCAACAAATCATAAAGATATTGG) and HCO2198 (reverse: TAAACTTCAGGGTGACCAAAAAATCA) (Folmer et al. 1994). The sequences obtained in this study were deposited into DDBJ (https://www. ddbj.nig.ac.jp/). The accession numbers of these sequences are listed in Table 1.

Phylogenetic analysis and BOLD Barcode Index Number clustering
To construct the phylogenetic tree, we downloaded the sequences of N. inferior, N. quadrimaculalis, and N. occultalis (two sequences, respectively) from GenBank.
Patania ruralis was included as the outgroup because Patania (= Pleuroptya) is considered to be closely related to Nagiella based on male and female genitalia (e.g., Munroe 1976;Inoue 1982), but wing maculation, host plants, and the results of the phylogenetic analysis of Lu and Du (2020) suggest they are clearly different (see also Discussion for the differences between Patania and Nagiella). The sequences were aligned using MEGA 7.0 (Kumar et al. 2016). A neighbor-joining (NJ) tree was constructed using MEGA 7.0 based on Kimura 2-parameter model (Kimura 1980), and the bootstrap values were calculated with 1,000 replicates. Detection of variation sites and the number of intra/interspecific substitutions were calculated using MEGA 7.0.
DNA barcoding employs DNA sequences in a short and standardized gene region to facilitate species identification. BOLD (http://www.boldsystems.org/) is an international repository of DNA barcodes (Ratnasingham and Hebert 2007). The sequences in BOLD are clustered depending on their divergences and each cluster is given a unique Barcode Index Number (BIN) (Ratnasingham and Hebert 2013), an identifier for DNA barcode-based cluster corresponding to species. We searched the BOLD database for BINs that matched sequences obtained in this study.

DNA sequence analysis
We successfully obtained 626 bp sequences of the COI barcode region of the seven specimens of Nagiella treated. Variation was detected at 58 sites (9.3%) in these 13 sequences. The number of intraspecific substitutions ranged from 0 to 3 (0-0.5%) while the number of interspecific substitutions ranged from 18 to 41 (2.9-6.5%) ( Table 2).  In the BOLD database, the sequence of N. quadrimaculalis obtained in this study corresponds to BOLD:AAD8178, and that of N. inferior corresponds to BOLD:AAE4571, while that of N. tristalis did not corresponded to any BIN.
The NJ tree ( Fig. 1) shows the four monophyletic clades that correspond to morphologically different Nagiella species with strong supports (bootstrap value of 100 for each species). The NJ tree also supports the close relationship between N. tristalis and N. inferior with moderate support (bootstrap value 77), and indicates N. occultalis as the sister species of N. quadrimaculalis with high support (bootstrap value 90).
2 Ground color of both wings lighter, postmedial line distinct especially in the hindwing; large, comma-shaped white spots at end of discal cell in each wing, usually larger; subdiscal white spot of forewing usually quadrilateral, distinct; base of discal cell of hindwing white; valva of male genitalia dorsally straight margined subapically; anterior apophysis of female genitalia slightly incurved to dorsally, expansion of the base sharply triangular; signum of female genitalia circular, small ( Etymology. The specific epithet refers to the darker wing color in comparison to that of N. inferior, and the habitat of this species is a shaded place. Diagnosis. This new species is similar to N. inferior and N. quadrimaculalis, also distributed in Japan, but it can be distinguished by the following characters: length of forewing 12.0-13.0 mm; vertex with erect, dull-orange scales; subdiscal white spot of forewing rounded, small, and blurry; base of discal cell of hindwing identical to ground color; dorsal margin of valva of male genitalia slightly incurved subapically; anterior apophysis of female genitalia straight and narrow; signum of female genitalia nearly elliptical, larger than in N. inferior (diameter 0.09-0.14 mm). This species is also similar to N. occultalis and N. bispina distributed in China, but N. occultalis has the following differences: subdiscal white spot of forewing narrowed or elongated, tuba analis of male genitalia sclerotized, gnathos of male genitalia elongated and narrow at the base; N. bispina exhibits the following differences: gnathos of male genitalia absent, phallus of male genitalia with a hook-shaped cornutus, corpus bursae of female genitalia with two thorn-like signa. From N. hortulatoides, the new species can be easily distinguished by the wing maculation. Description ( Fig. 2A, B). Head: frons brownish grey, smooth. Vertex with erect, dull-orange scales. Labial palpus upturned, dorso-laterally dark brown, ventro-mesally pale white. Antenna dark brown; flagellum filiform with golden cilia ca 1/4 the diameter of flagellum in male.
Wings: length of forewing 12.0-13.0 mm. Ground color of both wings brownish grey, with a large comma-shaped white spot at end of discal cell (the bases of R 5 to M 3 ), that of the hindwing somewhat small; cilia concolorous with ground color; postmedial line obscure. Subdiscal white spot of forewing rounded, small and blurry. Base of discal cell of hindwing concolorous with ground color.
Biology. In Honshu, Japan, adults are found in May to September, and they are considered bivoltine. They appear to be hardly attracted to light. Larvae feed on Rubus buergeri and the middle instar larvae overwinter in its leaves.
Host plant. Rubus buergeri Miq., R. palmatus Thunb. (in the field), R. buergeri, R. trifidus Thunb. (laboratory reared). Remarks. Nagia incomitata Swinhoe, 1894 has long been considered a synonym of N. quadrimaculalis, but based on the investigation of the type specimen, Lu and Du (2020) considered it likely to belong to Nosophora Lederer, 1863. We also follow this taxonomic treatment.
Our identification of this species in this study was based on external morphology (Kollar and Redtenbacher 1844;Inoue 1982;Li et al. 2012;Sasaki and Yamanaka 2013;Ullah et al. 2017;Lu and Du 2020) and male genitalia (Li et al. 2012;Ullah et al. 2017;Lu and Du 2020). The species status was confirmed by DNA barcodes.

Discussion
Recently, the integration of DNA barcoding and morphological approaches has accelerated various stages of taxonomic studies, such as species identification and description, re-investigation of taxa, as well as detecting cryptic species, also in Spilomelinae (Sutrisno 2005;Haines and Rubinoff 2012;Mally et al. 2016;Ullah et al. 2017;Ullah et al. 2018;Lu et al. 2019;Mally et al. 2019;Yang et al. 2019;Lu and Du 2020). In the genus Nagiella, these approaches led to the discovery of N. occultalis and N. bispina from China (Ullah et al. 2017;Lu and Du 2020), as well as N. tristalis from Japan (this study). Combined morphological and molecular biological studies might lead to the discovery of additional new species of this genus also in other regions.
The NJ tree shows N. inferior + N. tristalis and N. quadrimaculalis + N. occultalis to be sister groups (Fig. 1). The morphological evidence also supports these relationships, i.e., the gnathos of the male genitalia is short and triangular in N. inferior and N. tristalis (Figs 3B, 4B), while that of N. quadrimaculalis and N. occultalis is elongated (Fig. 5B); the signum of the female corpus bursae lacks projections in N. inferior and N. tristalis (Figs 6C, 7C), while that of N. quadrimaculalis has projections (Fig. 8C), although that of N. occultalis is unknown.
As the species of the genus Nagiella are very similar in appearance to each other (except for N. hortulatoides), DNA barcoding (see Material and methods) may provide very useful information for the identification of species in this genus. However, the species information for this genus in BOLD probably contains some misidentifications. For example, BOLD:AAD8178 cluster contains a record named "Pleuroptya inferior". This cluster can be identified as N. quadrimaculalis based on the results of this study and the specimen images in the deposited data. On the other hand, the sequences of N. inferior in this study corresponded to the BOLD:AAE4571 cluster, not BOLD:AAD8178. Therefore, "Pleuroptya inferior" in BOLD:AAD8178 is probably a misidentification. The user must judge whether the information in the database is based on correct identifications or not.
Host plant records of Rubus buergeri and R. sieboldii Blume for N. inferior, and R. buergeri for N. quadrimaculalis were known from Japan (Sasaki and Yamanaka 2013). However, in these published host plant records the Nagiella species may have been confused. In our laboratory, the three Nagiella species (including N. tristalis, reared from eggs) fed on R. buergeri and R. trifidus. In the field, we found larvae of N. quadrimaculalis feeding on R. buergeri and R. palmatus, and N. tristalis feeding on R. buergeri. In Tottori Prefecture, Japan, where the distribution of N. tristalis and N. inferior overlap, we could not find larvae of the latter species in the field, although where we found the larvae of the former species on R. buergeri in winter. This suggests that either 1) R. buergeri is not the native host plant of N. inferior, or that 2) N. inferior has a different overwintering strategy than N. tristalis. Contrary to our results, Fan and Piao (2013) recorded Rhus chinensis Mill. (Anacardiaceae) as a host plant for N. quadrimaculalis. The native host plants of the genus Nagiella needs further investigation.
The species of Nagiella have been placed in various genera, namely Coptobasis Lederer, Pleuroptya Meyrick, and Sylepta Hübner (e.g., Lederer 1863;Hampson 1899;Shibuya 1928;Inoue 1982;Irungbam et al. 2016). Munroe (1976) separated Nagiella from Pleuroptya and its related genera by the following characters: uncus short and wide, valva with a large oblique clasper, saccus relatively small and simple, and type of wing maculation (consisting of greyish ground color and a conspicuous white spot on each wing). Inoue (1982) placed N. inferior and N. quadrimaculalis in Pleuroptya, but no evidence for this treatment was provided. Inoue's opinion was followed by many authors (e.g., Li et al. 2012;Sasaki and Yamanaka 2013;Irungbam et al. 2016). Kirti and Gill (2007) treated Pleuroptya as a synonym for Patania Moore, because the male genitalia of their respective type species share congeneric characters as follows: valva leaf-like, uncus with a truncate posterior margin, gnathos absent, tuba analis elongate, and cornutus present in phallus. Ullah et al. (2017) regarded Nagiella as a valid genus based on the following characters: gnathos present, valva broader than that of Patania with stout subapical setae, and phallus without cornutus. However, several characters such as the presence or absence of a gnathos and the presence or absence of a cornutus in the phallus are shared with some Patania species. For example, P. clava Xu & Du, 2016 possesses a developed, finger-like gnathos (Xu and Du 2016), P. balteata (Fabricius, 1798) has an elongated gnathos (Leraut 2005;Slamka 2013), P. accipitralis (Walker, 1866) is missing cornuti in the phallus (Leraut 2005), and P. obfuscalis (Yamanaka, 1998) possesses a bunch of setae medially on the costa of the valva (Yamanaka 1998). In addition, the male genitalia of N. bispina described by Lu and Du (2020) lack the gnathos and possess a hook-shaped cornutus. Therefore, we tentatively regard Nagiella as a valid genus based on the following available characters: large oblique clasper, wing maculation, and host plants. Although in N. hortulatoides the wing maculation is different, this taxon is obviously included in Nagiella based on characters of the male genitalia as shown by the phylogenetic results of Lu and Du (2020). Although the separation of Nagiella has been accepted by many authors, such as Kirti and Sodhi (2001), Rose (2002), Mally et al. (2019), and Nuss et al. (2003Nuss et al. ( -2020, further comprehensive genitalic studies and also molecular phylogenetic analyses are indispensable to reveal details of the taxonomic status of Patania, Pleuroptya, and Nagiella.