Potential for interspecific hybridization between Zizina emelina and Zizina otis (Lepidoptera: Lycaenidae)
- 306 Downloads
Environmental changes such as global warming and biological invasion caused by human activities raise the possibility of secondary contact between the endangered butterfly species Zizina emelina and its sibling species Zizina otis in Japan. To assess the possible risks from their habitats overlapping, we investigated the potential for hybridization and the development of F1 individuals. We observed successful mating of the two sibling species under artificial conditions. The presence of a postzygotic hybridization barrier was supported by the delay of larval development only in females; a delay did not occur in males. Existence of the barrier was also supported by a decreased egg hatching rate in one brood; this was likely associated with infection with Wolbachia, a bacterium manipulating the reproductive capability of its host. The size and wing markings of F1 hybrid individuals were intermediate between those of the two species. These results suggest that, if Z. emelina and Z. otis are distributed sympatrically in the future, there is a possibility of introgression and reproductive interference between the two species, which would increase the risk of decline of each species.
KeywordsGeographic range expansion Interbreeding Introgression Postzygotic isolation Reproductive interference Wolbachia
Hybridization is a key phenomenon in the process of speciation and evolution (Barton and Hewitt 1985; Arnold 1997; Barton 2001). When species are formed in allopatry, hybridization can occur on the periphery of the species’ ranges if there is secondary contact (Arnold 1997; Seehausen 2004). Factors promoting secondary contact between species include environmental disturbances such as habitat alteration, global climate change, and biological invasion (Prentis et al. 2007; Nève et al. 2008; Crispo et al. 2011; Krehenwinkel and Tautz 2013). Such accidental hybridization can have potentially destructive outcomes via two main mechanisms (Wolf et al. 2001). First, if the hybrid is fertile and shows little reduction in fitness, the genetic purity of conspecifics is affected owing to genetic transfer via fertile hybrids (Wolf et al. 2001). Second, if the hybrid is infertile or exhibits reduced fitness compared with that of either species, the population growth rate in the hybridization zone may decline below that required for replacement (Wolf et al. 2001). Consequently, accidental hybridization threatens some animal and plant species with extinction (Rhymer and Simberloff 1996; Allendorf et al. 2001).
The lycaenid butterfly Zizina emelina emelina (de l’Orza) is distributed in Japan from the Kanto district southwestward to Tanegashima Island and is also found in some areas of Korea. This species inhabits sunny grasslands of early successional stages such as those of seashores, riverbanks, and farmland levees (Fukuda et al. 1984; Yago et al. 2008). Suitable habitats for Z. emelina are decreasing because of the loss of traditionally maintained farmlands and the use of concrete to cover river embankments (Yago et al. 2016). Consequently, the species is listed on the Japanese Red List of Insects as one of the most endangered species and is categorized as “Threatened I” (Ministry of Environment 2015). This butterfly, formerly known as Zizina otis emelina, is now treated as a species distinct from Zizina otis (Fabricius) according to the results of a study based on molecular, morphological, and host plant data (Yago et al. 2008).
Zizina emelina and Z. otis are distributed in temperate and tropical/subtropical zones, respectively. Their distributions in Japan were previously considered to be completely separate (Fig. 1a) (Yago et al. 2008). However, the range of Z. otis seems to be expanding northward, because it was recorded in 1995, 2008, and 2009 on Yakushima Island, which is very close to Tanegashima Island, the southern limit of Z. emelina (Fig. 1b) (Kubota 2009; Kawazoe 2012; Otsubo 2016). A great change in the distribution of butterflies in Japan has been reported (Inoue 2016), as represented by that of the Great Mormon butterfly, Papilio memnon, which has expanded its range in response to global warming (Yoshio and Ishii 1998; Kitahara et al. 2001). Butterflies and other insects show rapid responses to climate change, because they have short life cycles that are strongly influenced by temperature (Bale et al. 2002; Wilson and Maclean 2010; Caldas 2014). In recent decades, noticeable range expansion also has been reported in several other butterfly species [e.g., the Malayan butterfly, Megisba malaya; the lesser albatross, Appias paulina; and the common rose, Pachliopta aristolociae; (Fukuda 2012a, b)], especially from the Ryukyu Islands to the Kyushu mainland. This trend suggests that Z. otis may expand its range more widely toward the Kyushu mainland and that it may therefore share its habitat with Z. emelina. In addition, biological invasion is a major environmental problem globally (Vitousek et al. 1997; Mack et al. 2000; Clavero and Garcia-Berthou 2005). For example, Zizina otis labradus is considered to have been introduced into the natural habitat of Zizina oxleyi in New Zealand through anthropogenic activity (Gibbs 1980; Gillespie et al. 2013). Thus, human activity has the potential to result in secondary contact between Z. emelina and Z. otis.
In this study, we investigated the potential for hybridization between Z. emelina and Z. otis and the development of F1 individuals, and we discuss the potential effects if their ranges were to overlap in mainland Japan. Z. emelina is reported to be infected with Wolbachia, a bacterium that resides inside host cells and manipulates the reproductive capability of its host (Sakamoto et al. 2010, 2011, 2015b). One such manipulation, cytoplasmic incompatibility (CI), can act as a reproductive isolation barrier (Bordenstein et al. 2001). Therefore, we also investigated Wolbachia infection of mated individuals through crossing experiments.
Materials and methods
We used eight males of Z. otis field-collected from Ishigaki Island, Okinawa, on 16 July 2012, and 15 virgin females of second-generation progeny from a female Z. emelina field-collected in the city of Toyonaka, Osaka, on 12 May 2012. Under sunny conditions from 17 to 20 July, we placed one female and one to three males into a vertical cylindrical net (25 cm in diameter, 25 cm in height; Klark Co. Ltd., Aichi, Japan) and set up 3–5 of these nets simultaneously on the rooftop of a building at Osaka Prefecture University. Different combinations of males were placed in the net with each female at approximately 1-h intervals. Male approaches and the associated events were observed. Two pairs successfully mated. The mating pairs were picked up when the coupling had lasted more than 5 min, and they were maintained in the laboratory until mating ended. The two females were designated “T298-1” and “T298-2”, and the corresponding males were designated “Is1” and “Is2”, respectively.
Eggs were obtained and counted in the laboratory by offering the mated females leaves of bird’s-foot trefoil (Lotus japonicus) or white clover (Trifolium repens), two known host plants of Z. emelina (Fukuda et al. 1984; Minohara et al. 2007). The resulting broods were designated “Brood 1” and “Brood 2”. Sixty hybrid larvae randomly selected from each brood were reared on fresh leaves of T. repens until the second instar, then on an artificial diet from the third instar onward. For the artificial diet, dried leaf powder of T. repens was mixed with the powder of a ready-made diet material, Insecta F-II (Nihon-Nosan-Kogyo Co., Yokohama, Japan), using 30% dry weight of T. repens. Rearing experiments from egg to adult were carried out for up to 100 days at 25 °C under a photoperiod of 16:8 L:D. We examined the cumulative rates of larval and pupal molts. Pupal weight was measured on the fourth day after pupation by using an electric balance (Model HR-60, A&D Co. Ltd., Tokyo, Japan). The sex of each F1 hybrid was identified from the wing upperside color, and the morphological characteristics of all normally emerged F1 hybrids were evaluated.
Hatching and emergence rates were compared between broods by using Fisher’s exact test. Emerged male:female ratios of broods were evaluated by the binomial test. Differences in developmental periods at each stage and pupal weights between F1 males and females were analyzed by Student’s t test. All statistical tests were conducted using the statistical software R 3.1.1 (R Development Core Team 2014).
Female and male adults used in the crossing experiments were assayed to detect and identify the strain of any Wolbachia infection. Total DNA was extracted from the mid- and hindlegs. The legs were placed in a 1.5-mL plastic tube with 180 µL of lysis buffer and proteinase K, then incubated at 56 °C for at least 2 h, and finally subjected to DNA purification following the animal tissue protocol of the DNeasy Tissue Kit (Qiagen, Hilden, Germany). Total DNA was eluted with 50 µL of elution buffer. DNA quantification and qualification were performed with a V-550 UV/Vis spectrophotometer equipped with a SAH-769 one-drop measurement unit (JASCO, Tokyo, Japan). Detection and identification of Wolbachia were performed with two genes, Wolbachia surface protein (wsp) and Filamenting temperature-sensitive mutant Z (ftsZ). We used the primers 81F (5′-TGGTCCAATAAGTGATGAAGAAAC-3′) and 691R (5′-AAAAATTAAACGCTACTCCA-3′) for wsp (Braig et al. 1998) and the primers ftsZBf (5′-CCGATGCTCAAGCGTTAGAG-3′) and ftsZBr (5′-CCACTTAACTCTTTCGTTTG-3′) (Werren et al. 1995) for ftsZ. The PCR cycles for wsp were as follows: an initial 30-s exposure at 94 °C, followed by 40 cycles of 94 °C for 30 s, 55 °C for 30 s, and 75 °C for 120 s, with a final extension at 72 °C for 120 s. The PCR cycles for ftsZ were as follows: an initial 300-s exposure at 94 °C, followed by 45 cycles of 94 °C for 30 s, 60 °C for 60 s, and 72 °C for 120 s, with a final extension at 72 °C for 120 s. The positive PCR products were purified and sequenced with an ABI PRISM 3100 Genetic Analyzer (Applied Biosystems, Foster City, CA, USA). In addition, to confirm that DNA was properly extracted, the host mitochondrial cytoplasmic c oxidase I (COI) gene was amplified and sequenced using the primer set LCO1490 (5′-GGTCAACA AATCATAAAGATATTGG-3′) and HCO2198 (5′-TAAACTTCAGGGTGACCAAAAAATCA-3′) (Folmer et al. 1994) in the same samples. The PCR temperature profile for COI was an initial 120-s exposure at 94 °C, followed by 40 cycles at 94 °C for 15 s, 52 °C for 30 s, and 72 °C for 60 s, with a final extension at 72 °C for 300 s.
We observed successful mating of the two sibling species, Z. emelina and Z. otis. In some cases, couplings were broken within 10 s after mating seemed to have been established. The hatching rates were 78% (n = 251) and 100% (n = 120) in Broods 1 and 2, respectively, with a significantly different hatching rate between broods (p < 0.001, Fisher’s exact test). We examined the cumulative rates of larval and pupal molts in 60 individuals randomly selected from each brood (Fig. 2). Approximately half of the F1 individuals of both broods emerged within 40 days, but the other half progressively molted and began to emerge after 59 days. All emerged adults in the former group were males, whereas all emerged adults in the latter were females. The rates of emergence by 100 days differed significantly between Brood 1 (53%) and Brood 2 (90%; p < 0.001, Fisher’s exact test). The emerged male:female ratios of Broods 1 and 2 were 1:0.1 and 1:0.5, respectively; these were significantly different from the expected 1:1 ratio (p < 0.001, binomial test). If all individuals that did not emerge were assumed to be females, no biased sex ratios would be found in randomly selected hatched larvae of either brood (p > 0.05, binomial test). The larval period of F1 males was approximately four times shorter than that of F1 females (Table 1). The pupal period of F1 males was also shorter (Table 1). The male pupae weighed significantly less than female pupae (Table 1). Upon dissection of the abdomen of each emerged female, dozens of mature eggs were found.
Mean developmental periods of eggs, larvae, and pupae, and pupal weights of male and female F1 hybrids between Zizina emelina emelina and Zizina otis riukuensis
Number of samples
Mean developmental period ± SE (days)
Number of samples
Mean weight ± SE (mg)
4.4 ± 0.1
15.4 ± 0.2
7.8 ± 0.1
39.3 ± 0.6
4.5 ± 0.2
58.8 ± 1.7
9.1 ± 0.3
48.9 ± 1.9
Student’s t test
p > 0.05
p < 0.001
p < 0.001
p < 0.001
In external appearance, Z. emelina emelina differs from Z. otis riukuensis in having the following characteristics: (1) usually larger; (2) in males, marginal black dots on the hindwing upperside are distinct; and (3) black dots on the wing underside are much larger and more expanded, especially in submarginal lunules (Fig. 3). The size and wing markings of all F1 hybrid individuals were intermediate between those of Z. emelina emelina and Z. otis riukuensis (Fig. 3).
Two mated females (T298-1 and T298-2) and one mated male (Is1) were infected with Wolbachia, and the other male was not infected. The sequences of the two females were identical to those of wEmeTn1. The mated male of Z. otis, however, was infected with a novel strain of Wolbachia (referred to as “wOti1”). The length of the fragment of the wsp gene of wOti1, excluding primer sequences, was 555 nucleotides (deposited in the DDBJ/EMBL/GenBank databases under accession no. LC191397). In the ftsZ gene of wOti1, 911 nucleotide sequences were identified (accession no. LC191398).
Females of Z. emelina are physically able to mate with males of Z. otis under artificial conditions, indicating the possibility of successful interbreeding between the two species under natural conditions. If Z. otis invades the Kyushu mainland, the reproductive season of Z. otis may overlap temporally with that of Z. emelina and vice versa, because both species have five or more generations per year, and as a result adults fly continuously in summer and autumn (Fukuda et al. 1984; Sakamoto et al. 2015a). However, further work is required to examine in more detail the life cycles of the two Zizina species in each other’s climatic zones and assess whether the sequences of mating behavior match between them. In addition, in some cases, we noted uncoupling after mating seemed to have been established. Yago et al. (2008) noted differential morphology of the genitalia between Z. emelina and Z. otis. This difference is likely to act to some extent as a prezygotic barrier to mating.
The existence of a postzygotic hybridization barrier between these species was supported by the developmental delay. We found markedly slower larval development only in the F1 hybrid females (mean larval period: 58.8 days). However, the development time of F1 males (15.4 days) was similar to that of non-hybrid Z. emelina (18.5 days) (Sakamoto et al. 2015a). Hama (2012) also reported that no female offspring arose from mating between Z. emelina females and Z. otis males. Therefore, we conclude that only females have developmental issues in the F1 hybrids. This phenomenon can be explained by Haldane’s rule, which states that the heterogametic sex is more severely affected than the homogametic sex in the F1 hybrid generation (Haldane 1922; Coyne 1985; Futuyma 1986). In practice, only males emerge in interspecific F1 hybrids between various combinations of Papilionidae by hand pairing, because the female is the heterogametic (ZW) sex in lepidopterans (Ae 1986). The opposite cross would be expected to show a similar phenomenon, because lepidopterans exhibit heterogametic effects in reciprocal crossing (Haldane 1922; Ae 1986).
The existence of postzygotic isolation was also supported by the decreased egg hatching rate found in Brood 1, the father of which was infected with a strain of Wolbachia. Because the sex ratio of randomly selected larvae was presumed to be 1:1, one possible cause of the decreased hatching rate is CI induced by Wolbachia. CI causes embryonic mortality in crosses between infected males and uninfected females (or females infected with a different strain). The expression of CI ranges from very strong to very weak (Boyle et al. 1993; Poinsot et al. 2003; Yamada et al. 2007). Therefore, it is possible that the deaths in Brood 1 were associated with infection with the wOti1 strain of Wolbachia. Wolbachia-induced CI between populations can be an important factor in promoting speciation (Werren 1998; Bordenstein et al. 2001; Smith et al. 2016). Therefore, we intend to conduct more rearing experiments focusing on crossing Wolbachia strains to reveal the effects of wOti1. In addition, using a multilocus sequence typing scheme (Baldo et al. 2006) and more samples would improve our understanding of the infection history of Wolbachia and the speciation process of Zizina, both of which should be examined carefully.
Finally, we address the two main risks caused by the presence of an incomplete barrier between the two Zizina species. First, there is concern regarding introgression, because F1 males showed normal development and their intermediate morphology would be accepted by each species in copulation. The genetic purity of species is threatened by genetic transfer via such backcrossing (Ellstrand et al. 1999). If the two Zizina species’ distributions were to overlap, it would be necessary to analyze possible introgression. However, flow of mitochondrial DNA via maternal inheritance between these species is unlikely to occur because of the low viability of hybrid females, as Yago et al. (2008) and Cong et al. (2016) noted. Second, infertility and mortality diminish the fitness of a species, and some animal and plant species are reportedly exposed to the risk of extinction by such infertilities (Rhymer and Simberloff 1996), although a reproductive barrier can be an important driver of evolutionary events such as speciation, depending on the strength of the barrier (Cong et al. 2016). Reproductive interference is defined as any kind of interspecific interaction bringing about diminished fitness of at least one of the species in the mating process (Gröning and Hochkirch 2008). Kuno (1992) established a model describing the competitive interaction between two species in the mating process; they showed that those that interfere reproductively with each other essentially cannot coexist. The habitat range of a native species in New Zealand, Z. oxleyi, has been gradually decreased by expansion of the range of an introduced species, Z. otis labradus (Yago et al. 2008; Gillespie et al. 2013), probably because of reproductive interference via hybridization between the two species (Gillespie et al. 2013). The occurrence of a sympatric distribution between Z. emelina and Z. otis in the future may increase the risk of decline of each species.
We are grateful to Norio Hirai (Osaka Prefecture University [OPU]) for providing samples and kind assistance; to Kazuhisa Temma, Nao Temma, Takahiro Yoshida, Akito Inotsuka, Yoshihiro Torii, Haruka Matsuoka, and Sanae Mizuta (OPU) for assisting with the experiments; and to Minoru Ishii (OPU), Toshiya Hirowatari (Kyushu University), Koichi Goka (National Institute for Environmental Studies), Kazunari Nicho (The Lepidopterological Society of Japan), and Neil Moffat (Tokyo) for providing useful comments or documents. We also sincerely thank the staff of New Kansai International Airport Company, Ltd., for giving us opportunities to conduct the survey around the airport. This study was supported by a Grant-in-Aid for JSPS Fellows (No. 11J10420 to YS).
- Ae SA (1986) Biology of Papilionidae. Tatara Shobō, TottoriGoogle Scholar
- Arnold ML (1997) Natural hybridization and evolution. Oxford University Press, OxfordGoogle Scholar
- Bale JS, Masters GJ, Hodkinson ID, Awmack C, Bezemer TM, Brown VK, Butterfield J, Buse A, Coulson JC, Farrar J, Good JEG, Harrington R, Hartley S, Jones TH, Lindroth RL, Press MC, Symrnioudis I, Watt AD, Whittaker JB (2002) Herbivory in global climate change research: direct effects of rising temperature on insect herbivores. Glob Chang Biol 8:1–16CrossRefGoogle Scholar
- Caldas A (2014) Species traits of relevance for climate vulnerability and the prediction of phenological responses to climate change. J Lepid Soc 68:197–202Google Scholar
- Crispo E, Moore JS, Lee-Yaw JA, Gray SM, Haller BC (2011) Broken barriers: human-induced changes to gene flow and introgression in animals: an examination of the ways in which humans increase genetic exchange among populations and species and the consequences for biodiversity. Bioessays 33:508–518CrossRefPubMedGoogle Scholar
- Fukuda H (2012a) Northward range expansion in 1950–2010 in the Nansei Islands, southwestern Japan, reviewed for six butterfly species (2). Yadoriga (234):28–39 (in Japanese)Google Scholar
- Fukuda H (2012b) Northward range expansion in 1950–2010 in the Nansei Islands, southwestern Japan, reviewed for six butterfly species (1). Yadoriga (232):16–33 (in Japanese)Google Scholar
- Fukuda H, Hama E, Kuzuya T, Takahashi A, Takahashi M, Tanaka B, Tanaka H, Wakabayashi M, Watanabe Y (1984) The life histories of butterflies in Japan. III. Hoikusha, Osaka, Japan (in Japanese with English abstract)Google Scholar
- Futuyma DJ (1986) Evolutionary biology. Sinauer Associates, MassachusettsGoogle Scholar
- Gibbs GW (1980) New Zealand butterflies. Identification and natural history. Collins, AucklandGoogle Scholar
- Hama Y (2012) Studies of Ryukyuan butterflies I. Yuzuriha (52):26–31 (in Japanese)Google Scholar
- Inoue T (2016) Range expansion of butterflies due to various factors. In: Inoue T, Ishii M (eds). Range expansion of butterflies. The Hokuryukan Co., Ltd., Tokyo, pp 8–32Google Scholar
- Kawazoe N (2012) Zizina otis riukuensis in Yakushima Island. Gekkan-Mushi (498):41–42 (in Japanese)Google Scholar
- Kitahara M, Iriki M, Shimizu G (2001) On the relationship between the northward distributional expansion of the great mormon butterfly, Papilio memnon Linnaeus, and climatic warming in Japan. Trans Lepid Soc Japan 52:253–264Google Scholar
- Kubota Y (2009) Zizina otis riukuensis occurred in Yakushima Island. Satsuma (141):79–83 (in Japanese)Google Scholar
- Ministry of Environment, Japan (2015) The Japanese Red List 2015 on insects. http://www.env.go.jp/press/files/jp/28061.pdf. Accessed 1 Nov 2016
- Minohara S, Morichi S, Hirai N, Ishii M (2007) Distribution and seasonal occurrence of the lycaenid, Zizina emelina (de l’Orza) (Lepidoptera, Lycaenidae), around the Osaka International Airport, central Japan. Trans Lepidopterol Soc Jpn 58:421–432Google Scholar
- Otsubo S (2016) On Zizina emelina collected in Tanegashima Island. Satsuma (156):46 (in Japanese)Google Scholar
- R Development Core Team (2014) R: a language and environment for statistical computing. http://www.r-project.org/
- Sakamoto Y, Hirai N, Tanikawa T, Yago M, Ishii M (2011) Infection by two strains of Wolbachia and sex ratio distortion in a population of the endangered butterfly Zizina emelina (Lepidoptera: Lycaenidae) in northern Osaka Prefecture, Central Japan. Ann Entomol Soc Am 104:483–487CrossRefGoogle Scholar
- Werren JH (1998) Wolbachia and speciation. In: Howard DJ, Berlocher SL (eds) Endless forms: species and speciation. Oxford University Press, Oxford, pp 245–260Google Scholar
- Yago M, Hirai N, Kondo M, Tanikawa T, Ishii M, Wang M, Williams M, Ueshima R (2008) Molecular systematics and biogeography of the genus Zizina (Lepidoptera: Lycaenidae). Zootaxa 1746:15–38Google Scholar
- Yago M, Hirai N, Jinbo U (2016) The red lists of butterflies in prefectures, Japan—fourth edition (2015). In: Yago M, Hirai N, Jimbo U (eds). Decline and conservation of butterflies and moths in Japan. VII. The Lepidopterological Society of Japan, Tokyo, pp 83–351 (in Japanese)Google Scholar
- Yoshio M, Ishii M (1998) Geographical variation of pupal diapause in the great mormon butterfly, Papilio memnon L. (Lepidoptera: Papilionidae), in Western Japan. Appl Entomol Zool 33:281–288Google Scholar