Karyotype variation within the subgenus Stenocranius
The cytogenetic analysis shows that two cryptic species of narrow-headed voles, L. raddei and L. gregalis, are characterised by a number of karyotypic differences. L. gregalis has been subdivided into three allopatrically distributed and well genetically isolated lineages with unclear taxonomic rank. The main specific feature of L. gregalis karyotypes reported here is the presence of different numbers of Bs, which are not detectable in L. raddei karyotypes at all. The classification of size and morphology of Bs (Volobujev, 1981; Vujoševic & Blagojevic, 2004) suggested that Bs of L. gregalis belong to group I, i.e. are smaller than the smallest chromosome of the A chromosome set. Among examined B-carrying L. gregalis specimens, the size and morphology of Bs do not vary in our study; they are always present here as decreasing-in-size smallest acrocentric chromosomes of a karyotype. In contrast to some species with intra-individual variation of B chromosome numbers, for example, Leisler’s bat Nyctalus leisleri Kuhl, 1817 (Rajičić et al., 2022), all examined cells of each narrow-headed vole here contain a stable number of Bs. At the same time, B-carrying individuals are characterised by pronounced intra-population variation of Bs’ number.
The structure and chromosome morphology of the A chromosome complement do not vary among analysed L. gregalis karyotypes and match well those previously reported elsewhere. In both cryptic species, including all the genetic lineages of L. gregalis and L. raddei (for which a G-banded karyotype is presented here for the first time), the pattern of G-bands is similar and corresponds well to previously published ones (Lemskaya et al., 2010, Gladkikh et al., 2016). Except for differences in the localisation of C blocks on some chromosomes of the A complement and the presence of completely heterochromatic Bs in L. gregalis genetic lineages, C-band patterns are also similar between the two cryptic species.
Here, for the first time, we report data on differences in the numbers and localisation of rDNA clusters on Bs of L. gregalis. In a previous study, rDNA clusters were detected on six pairs of the A chromosome set (Gladkikh et al., 2016), but our results reveal that this pattern differs among all the genetic lineages of L. gregalis and L. raddei. Furthermore, an individual from above-mentioned studies seems to be a representative of lineage B, whereas karyotypes of voles of this lineage examined in the current study contain more (eight instead of six) rDNA clusters.
Behaviour, heterochromatisation and transcriptional inactivation of meio-Bs
Meiotic research on extra chromosomes in 25 mammalian species (data for 2018, see review Vujošević et al., 2018) and other vertebrates made it possible to determine the main features of the structure and behaviour of Bs. In prophase I, Bs were previously reported to be represented by different paired and unpaired meio-configurations (e.g. Świtoński et al., 1987; Kolomiets et al., 1988; Borbiev et al., 1990; Portela-Castro et al., 2000; Wang et al., 2000; Sosnowski et al., 2011; Karamysheva et al., 2017; Silva et al., 2021):
- univalents;
- univalents with self-synapsis, including formation of SYCP1-positive central elements;
- SC bivalents formed by chromosomes of the same length;
- SC bivalents formed by chromosomes of different lengths;
- multivalents (trivalents, tetravalents, etc. and trivalents with triple pairing).
In meiocytes of narrow-headed voles, only univalents are observed here. The number of B univalents does not vary in all spermatocytes and corresponds to mitotic karyotypes of the examined individuals. Thus, in contrast to other species, in Stenocranius voles, no interchromosomal differences in pairing behaviour of Bs are revealed in our work.
Stenocranius meio-Bs are heterochromatic according to our data. We confirm for the first time that H3K9me3 heterochromatin completely surrounds the unpaired axial elements of meio-Bs. Heterochromatic meio-Bs were noted in A. peninsulae (Bogdanov & Kolomiets, 2007). Those authors drew attention to the fact that likely signs of the heterochromatin nature of Bs are looseness or incompactness and sometimes branching of axial and lateral elements of an SC. To this set, one can add the irregularity of the SYCP3 signal distribution and the presence of irregular thickenings along the narrow-headed vole’s B univalents.
It has been proven that unpaired/unsaturated chromosomal sites undergo a process of transcriptional inactivation called meiotic silencing of unsynapsed chromatin (MSUC) (Schimenti, 2005; Baarends et al., 2005; Turner et al., 2005). The term ‘meiotic inactivation of sex chromosomes’ (MSCI) has been commonly used to refer to the inactivation of X and Y chromosomes as a special case of MSUC (McKee & Handel, 1993; Handel, 2004). It was previously found that if meiotic silencing affects chromosomal regions with critical genes, then this alteration may lead to pachytene checkpoint activation and to the demise of the meiocytes (Baarends et al., 2005; Turner et al., 2005). It has also been demonstrated that wide asynapsis and aberrant MSUC can cause impaired MSCI (Mahadevaiah et al., 2008; Burgoyne et al., 2009).
Because all narrow-headed voles’ prophase I supernumeraries are represented by univalents in our work, one could assume that such asynaptic elements are capable of disrupting meiotic progression. Nonetheless, we do not see features of such a process here. Most likely, the reason is that meio-Bs have usually been reported to be located together with asynaptic X and Y in the same chromatin domain, called the sex body, and therefore to undergo meiotic silencing. Such meiotic behaviour of Bs is reminiscent of an attempt to hide in an already existing chromatin ‘house’ (sex body) from close attention of the controlling pachytene checkpoint. It should be noted that this is the case when meiotic silencing of Bs (MSUC) and meiotic silencing of sex chromosomes (MSCI) are topologically located in the same chromatin compartment, although these processes are biochemically very similar and, in general, MSCI have been considered a type of MSUC (McKee & Handel, 1993; Handel, 2004).
The tendency towards co-localisation of Bs and sex chromosomes has been noticed previously in other species, for example, the silver fox Vulpes vulpes fulvus Desmarest, 1820 (Świtoński et al., 1987) and the Korean field mouse A. peninsulae (Kolomiets et al., 1988). The latter species has long been used as a good model for studying various aspects of extra chromosomes, including meiotic Bs. For instance, the inactivation of meio-Bs in this species was first reported by Ishak et al. (1991). Subsequently, it has been revealed that only some of A. peninsulae’sBs undergo transcriptional inactivation within the sex body, whereas the other Bs are located outside the sex body, and only some of them retain the γH2AFX signal (Karamysheva et al., 2017). This effect is probably due to the finding that some Bs were seen as univalents with self-pairing and bivalents did not have asynaptic regions in that study. Therefore, Apodemus and Stenocranius systems of meio-Bs differ from each other, although they manifested some signs of similarity.
Given that all Stenocranius Bs undergo transcriptional silencing and probably do not carry critical genes, additional asynaptic elements do not affect meiotic progression of spermatocytes. From an evolutionary point of view, such meiotic behaviour of Bs may contribute to their retention and circulation in populations of narrow-headed voles.
Polymorphism of supernumerary chromosomes in the evolutionary context
In the context of investigation into the mechanisms of population differentiation and subsequent speciation, here we focus on territories of the Altai-Sayan region and Transbaikalia, where several successive divergences within the subgenus Stenocranius probably have taken place. Furthermore, the zones of secondary contact between populations of the taxa in question are situated in these regions.
Based on dental morphology and molecular dating, it was proposed that L. raddei represents a relict of the Early Pleistocene, and that the Transbaikalia is the area of separation of two cryptic species of the subgenus Stenocranius (Petrova et al., 2016). All the newly examined karyotypes of L. raddei have a stable diploid chromosome number of 36, which is consistent with 2n = 36 in two previously reported karyotypes (Kovalskaya, 1989).
Lineage A of L. gregalis has been well researched cytologically, and a standard chromosome set (2n = 36) has been registered throughout its wide geographic range (Fedyk, 1970; Orlov et al., 1978; Lyapunova & Mirokhanov, 1969; Lemskaya et al., 2010; Sharshenalieva, 2014; Karipova, 2016; Tashibekova & Beshkempirova, 2019). Because variant 2n = 36 was found in the Tien-Shan Mountains (Fig. 1; Table 1), and these populations are the basal ones within lineage A as a whole (Petrova et al., 2015), the 2n = 36 chromosomal variant can be considered the initial one for L. gregalis lineage A. Our data are consistent with previous studies with the only exception: one specimen with Bs (2n = 36+3Bs) was found in Altai Mountains (Fig. 1, sampling site 4). It is difficult to guess the reasons for this finding, but it is curious that an individual with Bs was collected in the area of the probable separation of the three genetic lineages A, B and C. Perhaps an analysis of representatives of lineage A from the eastern part of the Tuva Basin and Southern slopes of the Western Tannu-Ola Range (Petrova et al., 2021) will resolve this question.
Among the previously analysed narrow-headed vole specimens, Bs were found in only two populations: in Khangai and Khentii Mountains in Central Mongolia; Kovalskaya (1989) proposed to explain this phenomenon by high seismic activity in this region. Our recent molecular genetic data indicated that these territories are inhabited by representatives of L. gregalis lineage B (Petrova et al., 2015). Our current analysis of additional karyotypic material from lineage B shows that Bs occur in vole populations sporadically (as far as it is possible to estimate by means of not very abundant material) albeit throughout the entire geographic range of this lineage: from South-Eastern Tuva to Transbaikalia. Prior to the present work, no cytogenetical investigations of representatives of L. gregalis lineage C, which inhabit a small territory of the western part of the Tuva Basin, have been carried out. We find that absolutely all examined voles of lineage C possess one to five additional Bs in karyotypes; not a single case of a standard set 2n = 36 is seen in our study.
Thus, all the obtained cytogenetical characteristics in combination with results of the molecular genetic analysis allow us to discern an interesting pattern. There is a possible way to construct an evolutionary series from older forms of L. raddei and L. gregalis lineage A (which normally do not have Bs) to younger lineages B and C, which carry one to five Bs. Moreover, lineage C apparently completely lost the karyotype variant having the initial chromosome set, 2n = 36, and demonstrates a lower proportion of karyotypes with 2n = 36+1B as well as a higher proportion of karyotypes with 2–4 Bs in comparison with lineage B (Fig. 2; Table S2).
The mechanisms of Bs’ origin in narrow-headed voles remain unclear. Nevertheless, our results indicate that specific features of C heterochromatin distribution on metaphase (mitotic) Bs and H3K9me3 heterochromatin on meio-Bs are very similar to these features in the Y chromosome. These characteristics, together with the localisation of meio-Bs and of X and Y chromosomes in the same chromatin domain, the sex body, suggest that Stenocranius extra chromosomes could have arisen from sex chromosomes. Further investigation is needed to test this hypothesis.
The finding that maximum diversity of Bs is located in the Altai-Sayan region is consistent with the long-proposed hypothesis about the influence of seismic activity on the occurrence of chromosomal rearrangements (Vorontsov & Lyapunova, 1984). It is worth mentioning that for A. peninsulae (Borisov & Zhigarev, 2018), the maximum number of Bs apparently is also observed in Southern Siberia. Perhaps, indeed, the reason is high seismic activity of the region and/or its high landscape heterogeneity, contributing to the spatial isolation of local populations and further speciation.
Rubtsov & Borisov (2018) proposed a hybridogenic nature of Bs with the caveat that traces of ancient hybridisation could already be eliminated from modern populations. There are no proven cases of such origin of Bs in mammals yet, but it has been documented experimentally in insects, Hymenoptera (Perfectti & Werren, 2001). The hypothesis about the possible hybridogenic nature of Bs is interesting in terms of narrow-headed voles because in the Altai-Sayan region, we see only a potential secondary contact zone of three young groups that have not yet acquired strict reproductive barriers (Petrova et al., 2016) but somehow maintain their genetic isolation (Petrova et al., 2021).
Further research is needed to clarify the origin, mechanisms of inheritance and stability of supernumerary chromosomes in populations of narrow-headed voles of the subgenus Stenocranius.