Karyotypes and COI Gene Sequences of Chironomus sp. Le1 (Kiknadze and Salova, 1996), Ch. laetus (Belyanina and Filinkova, 1996) and Their Hybrid from the Yamal Peninsula, Arctic Zone of Russia

Simple Summary Chironomids, or non-biting midges, are one of the most abundant insect families. Their larvae are an important component of various aquatic ecosystems. In the Chironomus genus, sibling species are found that are not distinguished by morphology. Due to the presence of giant polytene chromosomes in their salivary gland, it is more convenient to use cytogenetics for species identification. The molecular genetic method also has limitations and depends on the level of intra- and interspecific variability of species. We used morphological, cytogenetic, and molecular genetic approaches to the study of chironomid larvae from two lakes on the Yamal Peninsula. We found a few larvae, which by morphology were very similar to the Chironomus riihimakiensis group. By cytogenetics, we identified the species as Ch. laetus, the first species with arm combinations AE BC DF G and propose the name of a new cytocomplex—“laetus”. We also found a hybrid, Ch. laetus × Ch. sp. Le1, which is the first hybrid between species from different cytocomplexes. Molecular-genetic analysis of COI gene sequences has shown high values of genetic distance between Ch. laetus and species from the Ch. riihimakiensis group. Molecular genetic data suggests the presence of a subgroup with Ch. laetus. Abstract The study of the biological diversity of the Arctic zone yields intriguing results. Initial research on the lakes of the Yamal Peninsula resulted in the identification of Chironomus laetus and the hybrid Ch. laetus × Ch. sp. Le1. To avoid misidentification, we used morphological, cytogenetic, and molecular genetic approaches. By cytogenetics, in Ch. sp. Le1, seven banding sequences were found: Le1A1, Le1B1, Le1C1, Le1D1, Le1E1, Le1F1, and Le1G1. The karyotype of Ch. laetus was mapped for the first time; it is the first species with the arm combinations AE BC DF G. We propose the name of a new cytocomplex—“laetus”. DNA-barcoding of the COI gene was carried out for Ch. laetus and Ch. laetus × Ch. sp. Le1 for the first time. The estimated genetic distance between the sequences of Ch. laetus and Ch. riihimakiensis is 2.3–2.5%. The high similarity in morphology, banding sequences, and the possibility of hybridization indicate a close relationship between Ch. laetus and Ch. sp. Le1, which is assumed to be the northern variant of Ch. riihimakiensis. Molecular genetic data suggests the presence of a subgroup with Ch. laetus.


Introduction
The Arctic region has recently attracted the attention of biodiversity researchers. Special attention is dedicated to studying a species-rich family of chironomids, which play an important role in aquatic ecosystems.
Only a few species have been previously recorded from the region of investigation: Chironomus beljaninae (Wülker, 1991) [1], Ch. plumosus (Linnaeus, 1758), Ch. nigrifrons (Linevich and Erbaeva, 1971) [2], and also a lot of unidentified Chironomus spp., which were described only on the basis of larval morphology [3]. The vicinities of Vorkuta city (Figure 1), about 400 km from the studied location, is a type locality for three species: Chironomus laetus (Belyanina and Filinkova, 1996), Ch. frequentatus (Belyanina and Filinkova, 1996), and Ch. borealis (Belyanina and Filinkova, 1996), described without details of their karyotypes [4]. The main issue in the studies of Chironomus is the incorrect identification of species by morphological characteristics, especially at the larval stage, when reliable identification of some sibling species can be completely impossible. The most cost-effective and fastest method for the analysis of thousands and thousands of individuals in a short period is barcoding, which uses known sequences of DNA in any stage of the organism's development. Recently, this method was used to study the biodiversity of invertebrates in Canada [5], Norway [6], and South Korea [7]. The main disadvantage of this method is its low accuracy, which is associated with misidentification of species whose sequences have been deposited in genetic information databases (GenBank and BOLD) [8,9].
For example, for all Chironomidae registered in Germany, about 65% of the sequences do not have a species level of identification, the so-called "dark taxa" [10]. Previously, we found that the species name "Chironomus melanotus" matches several COI gene sequences, and the genetic p-distance between different populations varied from 0.4 to 12.4%, indicating that not all sequences belong to Ch. melanotus [11]. As we know, the genetic distance threshold is not suitable for all Chironomus species and depends on intra and interspecific sequence divergences; for the COI gene, it varies from 9 to 20% and, in rare cases, from 1 to 4% [12]. Due to the lack of sufficient information, we will use the 3% threshold suggested in the work [12][13][14][15][16].
The most accurate method of Chironomus identification is karyological analysis, which uses well-studied chromosomal banding sequences [17,18]. Many new cryptic species were described using this approach [18]. The main issue in the studies of Chironomus is the incorrect identification of species by morphological characteristics, especially at the larval stage, when reliable identification of some sibling species can be completely impossible. The most cost-effective and fastest method for the analysis of thousands and thousands of individuals in a short period is barcoding, which uses known sequences of DNA in any stage of the organism's development. Recently, this method was used to study the biodiversity of invertebrates in Canada [5], Norway [6], and South Korea [7]. The main disadvantage of this method is its low accuracy, which is associated with misidentification of species whose sequences have been deposited in genetic information databases (GenBank and BOLD) [8,9].
For example, for all Chironomidae registered in Germany, about 65% of the sequences do not have a species level of identification, the so-called "dark taxa" [10]. Previously, we found that the species name "Chironomus melanotus" matches several COI gene sequences, and the genetic p-distance between different populations varied from 0.4 to 12.4%, indicating that not all sequences belong to Ch. melanotus [11]. As we know, the genetic distance threshold is not suitable for all Chironomus species and depends on intra and interspecific sequence divergences; for the COI gene, it varies from 9 to 20% and, in rare cases, from 1 to 4% [12]. Due to the lack of sufficient information, we will use the 3% threshold suggested in the work [12][13][14][15][16].
The most accurate method of Chironomus identification is karyological analysis, which uses well-studied chromosomal banding sequences [17,18]. Many new cryptic species were described using this approach [18]. During analysis of the karyotypes of larvae from the Yamal Peninsula, we found a very strange chromosomal arm combination: "AB AE BC CD DF EF G", which consists of two different cytocomplexes. After thorough analysis, we concluded that it was an interspecies hybrid. One of the parents is Ch. sp. Le1, which belongs to the "thummi" cytocomplex [17], previously known from lakes in Yakutia [19].
Another parent is the Ch. laetus (Belyanina and Filinkova, 1996). Belyanina and Filinkova (1996) noted that Ch. laetus belong to the Chironomus "pseudothummi" cytocomplex, with the chromosomal arm combination AE BF CD G. However, analysis of its karyotype published in the aforementioned study showed that the authors have incorrectly identified chromosomal arms and the true combination was AE BC DF G; an unknown (new) cytocomplex.
In this work, we use a complex approach to the species identification of Chironomus. Preliminary investigation has shown the presence in GenBank and BOLD of about 50 unidentified COI gene sequences with high similarity to Ch. laetus, from Canada and Norway, and most of them situated outside of the Arctic Circle. The genetics of Ch. riihimakiensis have been studied insufficiently; only two COI gene sequences have been deposited in GenBank so far.
Chironomids are a very important component of aquatic ecosystems, and interest in them remains strong [7]. Molecular-genetic approaches are very useful in identifying the species at different stages of development, but this requires high accuracy in the first identification of species. The study's goal is to learn more about the Arctic distribution, morphological, cytological, and molecular genetic characteristics of Ch. laetus and the hybrid of Ch. sp. Le1 × Ch. laetus.

Materials and Methods
Fourth-instar larvae of Ch. laetus and the hybrid of Ch. laetus × Ch. sp. Le1 were collected from two lakes in the northeast Yamal Peninsula. Five larvae were found in Lake Pidarmato (71.253612, 71.638747) in June 2021. The depth of the lake is 3-4 m, the bottom sediments are silted sand, the pH is 6.8, and the dissolved oxygen concentration is 9.49 mg/L. One more larva was found in an unnamed lake (71.120760, 72.270375) in August 2021. The depth of the lake was 1-2 m; the bottom sediments are silted sand with inclusions of organic matter (rotting moss); a pH of 6.4; and dissolved oxygen concentration of 9.16 mg/L.
The head capsules of all larvae were mounted on a slide in the Fora-Berlese solution ( Figure 2). The morphological terminology proposed by Saether [30] was used. extended to the middle of the fourth segment, but sometimes it may be higher than the fifth segment; the ring organ is located at a length of 2/3 of the basal segment length (Figure 2a). Ventromental plates with flat frontal edges (Figure 2b). The fourth and fifth teeth of the mentum are in the same line as the third tooth and a little lower (Figure 2e). The fourth tooth of the mandible is small and lighter colored (Figure 2d). Lateral tubuli at segment VII are absent, while two pairs of ventral tubuli at segment VIII are present, with rounded apeces (Figure 2c).

Karyotypes of Chironomus from the Yamal Peninsula
The karyotype (half karyotype) of Ch. sp. Le1 was studied only in single hybrid larvae ( Figure 3). The chromosome set of the Ch. sp. Le1 is n = 4. The chromosome arm combination matches to the "thummi" cytocomplex: AB CD EF G. The chromosomes AB and CD are metacentric, EF is submetacentric, and G is telocentric. It was noted [17] that the arm G has the appearance of an amphora (Figure 3), which is typical for species of the Ch. riihimakiensis group. The bands in the centromeric region are heterochromatinized, and are not joined into the chromocenter [19,21]. The karyotype of the species is characterized by the presence of several nuclei and Balbiani rings, typical for the Ch. riihimakiensis group. The chromosome banding sequences of Ch. sp. Le1 are very similar to those of Ch. riihimakiensis, with the main differences being a larger (heterochromatinized) centromeric band and two dominant sequences, A2 and F2; it is thought that Ch. sp. Le1 is the northern variant of Ch. riihimakiensis [17,19,40].
Three larvae from the Pidarmato Lake and one from an unnamed lake, which were studied karyologically and morphologically, were used for the DNA extraction using the «M-sorb-OOM» (Sintol, Moscow region, Moscow, Russia) kit with magnet particles according to the manufacturer's protocol. For amplification of COI (cytochrome oxidase subunit I), primers LCO1490 (5 -GGTCAACAAATCATAAAGATATTGG-3 ) and HCO2198 (5 -TAAACTTCAGGGTGACCAAAAAATCA-3 ) were used (Eurogen, Moscow region, Moscow, Russia) [33]. The amplification reaction was carried out in a 25 µL reaction mixture (1× buffer, 1.5 µM MgCl 2 , 0.5 mM of each primer, 0.2 µM dNTP of each nucleotide, 17.55 µL of deionized water, 1 µL of template DNA, and 1 unit of Taq-polymerase (Evrogen, Moscow, Russia). PCR was performed at 94 • C for 3 min, followed by 30 cycles at 94 • C for 15 s, 50 • C for 45 s, 72 • C for 60 s, and a final cycle at 72 • C for 8 min. For the visualization of PCR products, 1% agarose gel was used, followed by purification with ethanol and ammonium acetate (3 M). Both strands were sequenced on an Applied Biosystems 3500 DNA sequencer (Thermo Scientific, Waltham, MA, USA) following the manufacturer's instructions.
For alignment of COI nucleotide sequences, we used MUSCLE in the MEGA6 software [34]. To calculate pairwise genetic distances using Kimura 2-parameter (K2P) with codon position preferences: first, second, third, and noncoding sites; the MEGA6 software (Pennsylvania State University, State College, PA, USA) was used [35]. The program MrBayes v.3.2.6 was used for the Bayesian analysis [36,37] with previously suggested settings [15,38] for 1,000,000 iterations and 1000 iterations of burn-in, with nst = 6 (GTR + I + G). The phylogenetic trees resulting from Bayesian inference analyses were visualized and edited using FigTree v. 1.4.3 [39].
In addition, twenty-nine COI gene sequences of the genus Chironomus from "GenBank" and "Barcode of Life Data Systems" (BOLD) were analyzed. Accession numbers of used sequences in GenBank and BOLD: Ch. acutiventris (

Morphological Characters of Ch. laetus from the Yamal Peninsula
The morphological characteristics of the fourth-instar larvae are presented in Figure 2. Most of the morphological characteristics correspond to the original description of Ch. laetus [4]. Body length is 10-14 mm. The head capsule is dark brownish in color. The frontoclypeus is dark with blurred borders. Occipital scleritis is dark. The antenna blade is extended to the middle of the fourth segment, but sometimes it may be higher than the fifth segment; the ring organ is located at a length of 2/3 of the basal segment length (Figure 2a). Ventromental plates with flat frontal edges (Figure 2b). The fourth and fifth teeth of the mentum are in the same line as the third tooth and a little lower (Figure 2e). The fourth tooth of the mandible is small and lighter colored (Figure 2d). Lateral tubuli at segment VII are absent, while two pairs of ventral tubuli at segment VIII are present, with rounded apeces (Figure 2c).

Karyotypes of Chironomus from the Yamal Peninsula
The karyotype (half karyotype) of Ch. sp. Le1 was studied only in single hybrid larvae ( Figure 3). The chromosome set of the Ch. sp. Le1 is n = 4. The chromosome arm combination matches to the "thummi" cytocomplex: AB CD EF G. The chromosomes AB and CD are metacentric, EF is submetacentric, and G is telocentric. It was noted [17] that the arm G has the appearance of an amphora (Figure 3), which is typical for species of the Ch. riihimakiensis group. The bands in the centromeric region are heterochromatinized, and are not joined into the chromocenter [19,21]. The karyotype of the species is characterized by the presence of several nuclei and Balbiani rings, typical for the Ch. riihimakiensis group. The chromosome banding sequences of Ch. sp. Le1 are very similar to those of Ch. riihimakiensis, with the main differences being a larger (heterochromatinized) centromeric band and two dominant sequences, A2 and F2; it is thought that Ch. sp. Le1 is the northern variant of Ch. riihimakiensis [17,19,40].

Karyotype of Ch. laetus
The karyotype of Ch. laetus has not previously been mapped. It was noted, that Ch. laetus by chromosome arm combination correspond to "pseudothummi" cytocomplex-AE CD BF G [4], but a valid arm combination is AE BC DF G and this is an unknown cytocomplex, which we propose to named "laetus". The chromosomes BC and DF are metacentric, AE is submetacentric, and G is telocentric (Figure 4). Visually we identified nucleoli (or puffs) in the arms D, F and G, which is typical for the Ch. riihimakiensis group [17].

Karyotype of Ch. laetus
The karyotype of Ch. laetus has not previously been mapped. It was noted, that Ch. laetus by chromosome arm combination correspond to "pseudothummi" cytocomplex-AE CD BF G [4], but a valid arm combination is AE BC DF G and this is an unknown cytocomplex, which we propose to named "laetus". The chromosomes BC and DF are metacentric, AE is submetacentric, and G is telocentric (Figure 4). Visually we identified nucleoli (or puffs) in the arms D, F and G, which is typical for the Ch. riihimakiensis group [17].

Genetic Distances Obtained with K2P
The pairwise genetic distances between the obtained sequences were calculated by the K2P model [35]. The distance between the hybrid of Ch. laetus × Ch. sp. Le1 and Ch. laetus sequences was 0.2%, and no differences were found between sequences of Ch. laetus from Pidarmato Lake and Unnamed Lake (Table 1).

Genetic Distances Obtained with K2P
The pairwise genetic distances between the obtained sequences were calculated by the K2P model [35]. The distance between the hybrid of Ch. laetus × Ch. sp. Le1 and Ch. laetus sequences was 0.2%, and no differences were found between sequences of Ch. laetus from Pidarmato Lake and Unnamed Lake (Table 1).   As previously shown, the studied species are members of the Ch. riihimakiensis group, and two COI gene sequences from Ch. riihimakiensis, MZ660756 and MZ659595, were found in GenBank; they were sampled in Finland, at the same locality (60 • 23 52.8 N and 23 • 05 56.4 E). The genetic distance between the sequences of Ch. riihimakiensis (MZ660756) and Ch. laetus was 2.5%, and the Ch. laetus × Ch. sp. Le1 hybrid was 2.3%. The genetic distance between sequences of Ch. riihimakiensis MZ660756 and MZ659595 was 12.4%, which is more than the accepted threshold of 3% [12][13][14][15][16]. The interesting point is that the distance between the sequences of Ch. laetus and species from the other members of the Ch. riihimakiensis group, Ch. novosibiricus and Ch. tuvanicus, is about 14%. One of the sequences of Ch. riihimakiensis (MZ659595) has a high degree of similarity with the sequence of Ch. tenuistylus (AF192201), approximately 3.1%.

The Analysis of the Phylogenetic Tree
The phylogenetic tree obtained by Bayesian inference showed groups of related species of Chironomus ( Figure 6). Here the situation is repeated (see Table 1). All sequences of Ch. laetus, including the hybrid, are combined into one cluster with Ch. riihimakiensis (MZ660756). Another sequence of Ch. riihimakiensis (MZ659595) forms a distinct cluster with Insects 2022, 13, 1112 9 of 13 the sequence of Ch. tenuistylus (AF192201). The two species from the true Ch. riihimakiensis group, Ch. novosibiricus and Ch. tuvanicus, form another distinct (independent) cluster.

The Analysis of the Phylogenetic Tree
The phylogenetic tree obtained by Bayesian inference showed groups of related species of Chironomus ( Figure 6). Here the situation is repeated (see Table 1). All sequences of Ch. laetus, including the hybrid, are combined into one cluster with Ch. riihimakiensis (MZ660756). Another sequence of Ch. riihimakiensis (MZ659595) forms a distinct cluster with the sequence of Ch. tenuistylus (AF192201). The two species from the true Ch. riihimakiensis group, Ch. novosibiricus and Ch. tuvanicus, form another distinct (independent) cluster.

Discussion
During the rare investigation of rivers and lakes located in the Arctic zone, every time it is possible to make interesting findings about chironomids (a new species, banding sequences, etc.) [4,15,21,42].
Chironomids are one of the essential components in most bodies of water. Because of this, they are convenient objects for environmental monitoring and the study of biodiversity. The main obstacle to this is the difficulty of sibling-species identification using only larval [18,19], pupal, and imaginal morphology. In order to avoid misidentification, it is suggested to use a comprehensive approach including cytogenetic and molecular-genetic analysis [8,12,13].
In Pidarmato Lake and an unnamed lake in the Yamal Peninsula, we collected larvae of Chironomus with a dark brown head capsule. They possess the morphological characteristics that are typical for the Chironomus riihimakiensis group, and at first, we were not sure of the species identification accuracy.
Cytogenetic analysis of one larva revealed a very interesting karyotype (chromosome set), which consisted of two different species' chromosomes. It was a hybrid of Ch. sp. Le1, a northern variant of Ch. riihimakiensis [17,40], and another species from the unknown cytocomplex, so we thought that the species was also new. In one paper devoted to the study of three sympatric species of Chironomus, we found the figure with the very similar karyotype of Ch. laetus [4].
The karyotype of Ch. laetus from the Yamal Peninsula is very similar to the first photomap presented by Belyanina and Filinkova on Figure 6a in [4]. However, the authors indicated the wrong "pseudothummi" cytocomplex with the chromosome arm combination AE CD BF G [4]. They have mixed up the arms of C and F in CD and BF combinations. It would be correct to specify combinations of BC and DF. Visually, we noted differences only in the absence of nucleoli in arm E of our specimens. This may be an effect of the environmental conditions. In Chironomus species evolution, of great significance is the reciprocal translocation of whole chromosomal arms and, as a result, the formation of diverse arm combinations. Based on this, all Chironomus species are grouped into 17 different cytocomplexes [19,43]. We propose the name "laetus" for the new cytocomplex because Ch. laetus is the first species with the arm combinations AE BC DF G.
We performed a preliminary analysis of the banding sequences of Ch. laetus and found their similarity with the already known sequences of the primitive Chironomus riihimakiensis group. The most conservative banding sequences in Chironomus are located in arms E and F [31,44], and this species is no exception. The laeE1 banding sequence is very similar to most species from the cytocomplexes «thummi» and «pseudothummi», including Le1E1, Al1E1, and pluE1. The laeF1 banding sequence is very similar to sorF1, abeF1, and borF1; note that the part of the sequence "17d-11a 18a-23f" excluding the telomere region, is characteristic of most of the species Chironomus [43]. The banding sequence laeA1 is very similar to Le1A1, Al1A1, and a few species from the cytocomplexes "thummi" and "pseudothummi". The sequences in arm B are more species-specific [45]. With reference to laeB1, only with the Al1B2 sequence we did find a high degree of similarity. The banding sequence laeC1 is very similar to Al1C1, pilC1, sorC2, and abeC1. The banding sequence laeD1 is similar to Tu1D2.
The occurrence of interspecific hybridization in chironomids is a quite rare event [23][24][25]27,28]. Our finding of a hybrid between Chironomus from different cytocomplexes is presented for the first time. This means that there is no complete reproductive isolation between two closely related species. The new cytocomplex (and the species) are probably very young, as Ch. laetus is not only close to other species in the Ch. riihimakiensis group but can also form hybrids. This is confirmed by the conjugation of homologues in the telomere regions of the two most conservative arms, E and F. The hybrid species' chromosome sizes are very similar; the morphological characteristics of centromeric bands, nucleoli, and Balbiani rings position correspond to species-specific features ( Figure 5).
We have no opportunity to analyze the COI gene sequence of Ch. sp. Le1, because no larvae of this species were found to be hybrid. The hybrid (OP205478) and Ch. laetus sequences (OP205477, OP199059, and OP205321) differ by 0.2% or one nucleotide. We can assume that the COI gene sequence was from Ch. Sp. Le1, the maternal organism. Otherwise, it might be the result of nucleotide polymorphism; a similar case was noted in Ch. bonus when each of the three sequences had a different haplotype [16].
During the comparison of the genetic distance between the sequences of Ch. laetus and Ch. riihimakiensis (MZ660756 and MZ659595) from Finland, we received two different values, which are 2.5 and 13.4%. The distance between the two sequences of Ch. riihimakiensis was 12.4%, which is much more than the 3% accepted interspecific threshold [12,13]. It could be a result of the misidentification by morphological characteristics of one of them or both [8,9]. We can assume that they are two different species. We have not found any sequences of the species from the Ch. riihimakiensis group, and existing data is not enough to make final conclusions about the status of the species on the evolutionary tree. Nevertheless, analysis of one short segment of the COI gene is not enough to make final inferences [8]. The high degree of similarity between Ch. laetus banding sequences, larval morphological features, and the event of hybridization with the Ch. riihimakiensis group suggests a close relationship. At first, grouping was subjective (artificial) and had no precise criteria, so the species were united by some similarity of morphology and karyotype [43]. The position of Ch. laetus on the Bayesian tree in separate clusters from Ch. riihimakiensis group suggests the presence of the molecular-genetic subgroup of the species.
We cannot ignore the presence of 48 COI gene sequences of Chironomus sp. with a 99-100% match with Ch. laetus in the GenBank, for example, Ch. sp. MN676646.1 (Canada, Cambridge Bay, 69 • 13 09.8 N, 104 • 55 12.0 W). This is a great opportunity to perform a preliminary investigation of the Arctic zone's biodiversity with subsequent extension to morphological and cytogenetic examination techniques.
Author Contributions: Conceptualization, V.B., A.P. and E.M.; sample collection, D.P., A.A. and A.P.; conceptualization and experimental-molecular analyses, project administration, karyological and statistical analysis, and manuscript-writing, V.B.; manuscript-writing (input), A.P.; contribute to the study design and manuscript editing E.M., D.P. and A.A. All authors have read and agreed to the published version of the manuscript.