Chromosomal and DNA barcode analysis of the Melitaea ala Staudinger, 1881 species complex (Lepidoptera, Nymphalidae)

Abstract The species of the Melitaea ala Staudinger, 1881 complex are distributed in Central Asia. Here we show that this complex is a monophyletic group including the species, M. ala, M. kotshubeji Sheljuzhko, 1929 and M. enarea Fruhstorfer, 1917. The haploid chromosome number n=29 is found in M. ala and M. kotshubeji and is, most likely, a symplesiomorphy of the M. ala complex. We show that M. ala consists of four subspecies: M. ala zaisana Lukhtanov, 1999 (=M. ala irtyshica Lukhtanov, 1999, syn. nov.) (South Altai, Zaisan Lake valley), M. ala ala (Dzhungarian Alatau), M. ala bicolor Seitz, 1908 (North, East, Central and West Tian-Shan) and M. ala determinata Bryk, 1940 (described from “Fu-Shu-Shi”, China). We demonstrate that M. kotshubeji kotshubeji (Peter the Great Mts in Tajikistan) and M. kotshubeji bundeli Kolesnichenko, 1999 (Alai Mts in Tajikistan and Kyrgyzstan) are distinct taxa despite their geographic proximity in East Tajikistan. Melitaea enarea is widely distributed in the southern part of Central Asia and is sympatric with M. kotshubeji.


Introduction
This work is a continuation of a series of publications (Lukhtanov and Kuznetsova 1989;Pazhenkova et al. 2015;Pazhenkova and Lukhtanov 2016;Lukhtanov 2017) devoted to the analysis of chromosomal and mitochondrial haplotype diversity and taxonomy of butterflies of the species-rich butterfly genus Melitaea Fabricius, 1807. The combination of molecular and cytogenetic methods is a useful tool for taxonomic studies Pazhenkova and Lukhtanov 2019) and can be a good addition to morphological analysis of taxonomically complicated groups of species . In our previous papers, we applied analysis of the DNA barcodes and karyotypes to study the genetic and taxonomic structure of the M. didyma (Esper, 1779) (Pazhenkova et al. 2015;Pazhenkova and Lukhtanov 2016) and M. persea Kollar, 1849 (Lukhtanov 2017) species complexes. The aim of this work is to study a complex of species close to M. ala Staudinger, 1881.
Molecular phylogenetic analysis (Leneveu et al. 2009) demonstrated that M. ala and M. enarea (cited in the article as M. permuta Higgins, 1941) are sister species, and M. acraeina is a phylogenetically distant species which is a sister to the lineage (M. ala + M. enarea). Melitaea sutschana was found as a member of the M. didyma species complex which is a sister to the lineage ((M. acraeina + (M. ala + M. enarea)) (Leneveu et al. 2009). In our study, we focused on the analysis of the M. ala lineage. We did not include M. ninae, M. didymina and M. chitralensis in the analysis, since for these species there has been no material suitable for chromosomal and molecular studies.

Chromosomal analysis
Karyotypes of four samples of M. kotshubeji kotshubeji were studied as previously described Vishnevskaya et al. 2016). Briefly, gonads were removed from the adult male abdomen and placed into freshly prepared fixative (3:1; 96% ethanol and glacial acetic acid) directly after capturing the butterfly in the field. Testes were stored in the fixative for 3-36 months at +4 °C. Then the gonads were stained in 2% acetic orcein for 5-10 days at +18-20 °C. Different stages of male meiosis, including metaphase I (MI) and metaphase II (MII) were examined using an original two-phase method of chromosome analysis (Lukhtanov et al. 2006(Lukhtanov et al. , 2008. Leica DM2500 light microscope equipped with HC PL APO 100×/1.44 Oil CORR CS lens and S1/1.4 oil condenser head was used for bright-field microscopy analysis. A Leica DM2500 light microscope equipped with HC PL APO 100×/1.40 OIL PH3 lens was used for phase-contrast microscopy analysis.

Molecular methods and DNA barcode analysis
Standard COI barcodes (658-bp 5' fragment of mitochondrial cytochrome oxidase subunit I) were studied as previously described Vishnevskaya et al. 2016). COI sequences were obtained from 34 specimens representing the M. ala species group and outgroups (M. telona Fruhstorfer, 1908 andM. alatauica Staudinger, 1881). Legs were used as a source for DNA isolation Legs from 6 specimens (M. kotshubeji bundeli Kolesnichenko, 1999) were processed in the Department of Karyosystematics of Zoological Institute of the Russian Academy of Sciences using primers and protocols described by Shapoval et al. (2017). Sequencing was carried out at the Research Resource Center for Molecular and Cell Technologies of St. Petersburg State University.
Legs from 28 specimens of Melitaea spp. were processed in the the Canadian Centre for DNA Barcoding (CCDB, Biodiversity Institute of Ontario, University of Guelph) using their standard high-throughput protocol described by Hajibabaei et al. (2005), Ivanova et al. (2006) anddeWaard et al. (2008). The set of voucher specimens of butterflies is kept in the Zoological Institute of the Russian Academy of Science (St. Petersburg) and in the McGuire Center for Lepidoptera and Biodiversity (MGCL), Florida Museum of Natural History, University of Florida, Gainesville, Florida, USA. Photographs of these specimens, as well as collecting data are available in the of Life Data System (BOLD), projects Butterflies of Palearctic (BPAL) and Butterflies of Palearctic Part B (BPALB) at http://www.boldsystems.org/.
Sequences were aligned using the BioEdit software (Hall 1999) and edited manually. Phylogenetic hypotheses were inferred using Bayesian inference as described previously (Vershinina and Lukhtanov 2010;Przybyłowicz et al. 2014;Lukhtanov et al. 2016). Briefly, the Bayesian analysis was performed using the program MrBayes 3.2 (Ronquist et al. 2012) with default settings. Two runs of 10,000,000 generations with four chains (one cold and three heated) were performed. We checked runs for convergence and proper sampling of parameters [effective sample size (ESS) > 200] using the program Tracer v1.7.1 (Rambaut et al. 2018). The first 25% of each run was discarded as burn-in. The consensus of the obtained trees was visualized using FigTree 1.3.1 (http://tree.bio.ed.ac.uk/software/figtree/).

Karyotype
The haploid chromosome number n=29 was found in prometaphase I, MI and MII cells of four studied individuals of M. kotshubeji kotshubeji (Table 2, Fig. 1). All chromosome elements formed a gradient size row. The karyotype contained no exceptionally large or small chromosomes.

DNA barcode analysis
DNA barcode analysis revealed M. ala, M. kotshubeji and M. enarea as highly supported monophyletic entities. Together, these three species formed a monophyletic lineage (the M. ala species complex) (1 in Fig. 2). In relation to the M. ala species complex, M. acraeina was found as a phylogenetically distant sister group (2 in Fig. 2). Taxa close to M. didyma (the M. didyma species complex) also formed a clade, but its support was relatively low (3 in Fig. 2). The species M. deserticola formed an independent lineage within the M. didyma species group (4 in Fig. 2). Together, these four lineages (M. ala complex + M. acraeina + M. didyma complex+ M. deserticola) formed the well-supported M. didyma species group (I in Fig. 2). The species of the M. persea group also formed a supported clade, sister to the M. didyma group (5 and II in Fig. 2). Within the M. ala clade, five supported (Bayesian posterior probabilities ranged from 0.9 to 1.0), relatively weakly differentiated subclades were found. These are (1) M. ala ala, (2) M. ala irtyshica, (3) M. ala zaisana, (4) M. ala bicolor (clade b1) and (5) M. ala bicolor (clade b2). We also calculated the uncorrected COI p-distances within (Table 3)

and between (Table 4) the revealed clades and groups.
Melitaea kotshubeji kotshubeji and M. kotshubeji bundeli were found to differ by four fixed nucleotide substitutions in the COI barcode region.

Chromosome number variation
The genus Melitaea (Fabricius, 1807) has relatively low interspecific chromosome number variation. The representatives of basal clades (see phylogeny in Leneveu et al. 2009 (Federley 1938;de Lesse 1960;Robinson 1971;Larsen 1975;Hesselbarth et al. 1995). These haploid   numbers are modal ones not only for Melitaea, but also for the family Nymphalidae and for the order Lepidoptera in whole (Robinson 1971;Lukhtanov 2000Lukhtanov , 2014. Most likely, one of them (probably, n=31, see Lukhtanov 2014) represents an ancestral lepidopteran state preserved in the basal lineages of Melitaea.
The Melitaea didyma species group is one of the younger lineages of Melitaea (Leneveu et al. 2009). This group is found to have lower chromosome numbers varying from n=27 to n=29-30 (Table 5). Melitaea didyma species complex is characterized by chromosome numbers from n n=27 to n=30, with n=28 and n=29 as modal numbers. In the Melitaea deserticola species complex, only one species (M. deserticola) is karyotyped (n=29). In the Melitaea persea species complex, n=27 is found in two species. In the Melitaea ala species complex, n=29 is found in two species studied.
Based on the distribution of the known chromosome numbers (Table 3) relative to the phylogeny (Fig. 2) and on the frequency of their occurrence, we can assume that n=29 is an ancestral state for the species of the M. didyma group. Thus, for the species of the M. ala complex n=29 is a symplesiomorphy.

Intraspecific taxonomy of the M. ala species group
The five identified clades within the species M. ala have relatively high support (Fig.  2) and can be considered as taxa, at least from the standpoint of the phylogenetic species concept (Cracraft 1989;Coyne and Orr 2004), in which diagnosable entities can be classified as species regardless of whether there is reproductive isolation between them or not. To assess the possibility of interpreting these clades as species or subspecies, we compared the level of COI divergence between the clades with the level of variability within the clades (Tables 3, 4). We found that in all cases, the distances between these clades were lower than 'standard' DNA-barcode species threshold (3%) (Hebert et al. 2003).
An especially low level of differentiation (0.3-0.5%) was found between the clades M. ala zaisanica and M. ala irtyshica. Therefore, we are inclined, especially taking into account the geographical proximity of these lineages, to consider them as a single taxonomic unit, M. ala zaisanica (= M. ala irtyshica).
A slightly higher average level of differentiation (0.3-0.8%) was found between the b1 and b2 clades (Fig. 2, Table 4). However, in this case, a rather high level of intragroup variability was observed (Table 3), and the maximum values of intragroup variability exceeded the minimum intergroup differences. Therefore, taking into account the geographical proximity of these lineages, we decided to consider them as a single taxonomic unit, M. ala bicolor.
Thus, within the studied populations, three subspecies can be distinguished. These are M. ala ala, M. ala bicolor and M. ala zaisana.
Melitaea ala ala is distributed in the Dzhungarian Alatau in East Kazakhstan (Fig. 3). This subspecies is characterized by darkening of the veins on the underside of the hind wing. These darkened veins form clear cells in the region of the median band (Fig. 4a).
Melitaea ala bicolor Seitz, 1908 is distributed in the North, East, Central and West Tian-Shan in SE Kazakhstan, NW China and Kyrgyzstan (Fig. 3). In this subspecies the veins on the underside of the hind wing are not strongly darkened. The cells of the median band are not highlighted. They are only marked with dark brackets on the outside of the median band (Fig. 4b). The specimens from the Tyshkantau Mts (SE part of the Dzhungarian Alatau in Kazakhstan) (Tuzov and Churkin 2000) and the eastern most part of the Tian-Shan (Kolesnichenko 1999)

are intermediate between M. ala ala and M. ala bicolor.
With regards to DNA barcodes, M. ala zaisana Lukhtanov, 1999 (Fig. 4c) is distinct from the geographically closest M. ala ala. With regards to the wing pattern, M. ala zaisana is more similar to M. ala bicolor than to M. ala ala. Interestingly, the northernmost population of M. ala from Oktyabrsk (Kazakhstan) (Fig. 3d) is intermediate in its appearance between M. ala ala and M. ala zaisana. This population was described as M. ala irtyshica Lukhtanov, 1999(Lukhtanov 1999 and was later erroneously synonymized with M. latonigena Eversmann, 1847 (Lukhtanov et al. 2007). DNA barcode analysis demonstrates that this population is similar to M. ala zaisana.
Currently, there is a tendency to consider as a species any group of populations with a minimum set of fixed differences. We are almost certain that, given this trend, the subspecies discussed above will be interpreted by some authors as species in the future. Nevertheless, in our opinion, in accordance with the subspecies criteria De Queiroz, 2020), they should be treated as subspecies of the same species.
Melitaea kotshubeji bundeli (Fig. 4h, i) was described as subspecies of Melitaea kotshubeji (Fig. 4j) (Kolesnichenko 1999), but later was treated as a distinct species (van Oorschot and Coutsis 2014) or alternatively as a synonym (Tshikolovets 2003(Tshikolovets , 2005. Our study demonstrates that these two taxa are not only distinct in the wing pattern, but also differ by four fixed nucleotide substitutions in the DNA barcode region, indicating the relative long independent evolution of these two sublineages. Interestingly, the distribution areas of these two allopatric taxa are in close proximity to each other and are separated by a narrow valley of the Surkhob River (in Kyrgyzstan, this river is called the Kyzylsu).
In our work we do not consider the intraspecific structure of M. enarea (Fig. 4k, l) due to the lack of molecular data for the northern populations of this species. The taxa described by Bryk (1940) Bryk (1940 described four taxa (all as subspecies of M. didyma) that should be assigned to M. ala. The types of these taxa were studied by the first author of this article in 2007 during a visit to Swedish Museum of Natural History.
The taxon described by Bryk (1940) as M. didyma allah Bryk, 1940 has the wing pattern with clear characters of M. ala ala (Fig. 5a, b), but not of the subspecies M. ala zaisana (Fig. 4c) as supposed by Tuzov and Churkin (2000). Thus, M. didyma allah should be synonymized with M. ala ala as suggested by Kolesnichenko (1999). We agree with Kolesnichenko (1999) that the label data of the syntype of M. didyma allah (Fig. 5c) are probably wrong.  The taxa described by Bryk (1940) as M. didyma sheljuzhkoi Bryk, 1940 (Fig. 5g-i) and M. didyma strandi (Fig. 5j-l) have the wing pattern with characters of M. ala bicolor. Most likely, they represent synonyms of M. ala bicolor.
The taxon from "Fu-Shu-Shi" (China) described by Bryk (1940) as M. didyma determinata Bryk, 1940 is characterized by the well-developed black wing pattern on both wing upper-and underside (Fig. 5d-f ). Most likely, it represents a distinct subspecies. Unfortunately, we do not have material for molecular study to test this hypothesis.