A Comprehensive Characterization of the Honeybees in Siberia (Russia)

A comprehensive study of some populations of honeybee (332 colonies) in Siberia (Tomsk region, Krasnoyarsk Krai (Yenisei population), Altai) using morphometric and molecu‐ lar genetic methods was conducted. Infestation of bees (132 colonies) by Nosema has also been studied. Three variants of the COI-COII mtDNA locus were registered: PQQ, PQQQ (typical for Apis m. mellifera), and Q (specific for southern races). It was established that 64% of bee colonies from the Tomsk region and all colonies studied from the Kras‐ noyarsk and the Altai territories originate from Apis m. mellifera on the maternal line. According to the morphometric study, the majority of bee colonies of the Tomsk region are hybrids; in some colonies the mismatch of morphometric and mtDNA data was observed. Moreover, the majority of bee colonies infected by Nosema were hybrids. Yenisei population may be considered as a unique Apis m. mellifera population. Microsatellite analysis (loci А008, Ap049, AC117, AC216, Ap243, H110, A024, A113) showed the specific distribution of genotypes and alleles for some loci in the bees, which differ by geograph‐ ical location. Loci A024 and Ap049 are of considerable interest for further study as candidate markers for differentiation of subspecies; locus A008 can be considered informative for determining of different ecotypes of Apis m. mellifera.


Introduction
In Siberia, the honeybee was introduced about 230 years ago. It was the dark-colored forest bee Apis mellifera mellifera L., or the Middle Russian race (a term adopted in Russia), that was cultivated in Siberia as the most adapted to the harsh climatic conditions of the region. At the end of the last century, bees of southern races, such as the Carpathian race or Apis mellifera carpatica (a derivative of A. m. carnica) and the Caucasian gray mountain race (Apis mellifera caucasica Gorb.), have been actively imported to Siberia. This process had become widespread and almost uncontrollable, which leads to a high level of crossbreeding of bees.
At present, one of the beekeeping problems in different countries is a massive bee hybridization, which leads to the reduction of the range of native subspecies, the formation of hybrids, and "deterioration" of the genotypic composition of honeybees. Hybrid populations are less adapted to environmental conditions that rapidly change during the year and are characterized by the higher morbidity and low immunity [1][2][3].
Introgressive hybridization modifies the genetic pool of local honeybee populations leading to the loss of their genetic identity [4]. The process of hybridization of different subspecies of honeybee can cause the destruction of the established gene complexes, leading to decrease in adaptive properties of organisms and populations and the change in biological and economically significant indicators of bees. The observed widespread hybridization of honeybees and the formation of hybrid bees can certainly contribute to the spread of disease. The extent of hybridization, characteristics of hybrid bees, the study of genetic processes that occur during hybridization, and evaluation of the effects of hybridization are of considerable interest.
The goal of this study is the morphometric and molecular genetic (mtDNA and microsatellite analysis) characterization of honeybees in Siberia and the assessment of the infestation of bee colonies by Nosema.

Region
Bees and bee colonies were investigated in three regions of Siberia: the Tomsk region, the Krasnoyarsk Krai, and the Altai Krai (Figure 1).
The Tomsk region is located in the geographic center of Siberia, in the southeastern part of the West Siberian Plain. The distance between the northern and southern boundaries of the meridian is about 600 kilometers; therefore, the climate of the southern and northern regions is markedly different. A climatic characteristic of the northern region is a more severe and prolonged winter season. Almost the entire territory of the region is within the taiga zone, where forests cover about 60% of the territory. The climate is temperate continental with considerable daily and annual amplitudes and long winters (5-6 months). The average annual temperature is -0.6 °C, while the average temperature in July is +18.1 °C and in January is 19.2 °C. The frost-free period is 100-105 days. Precipitation is 435 mm.  The Krasnoyarsk Krai is located in the Eastern Siberia. The climate is sharply continental, where 70% of the territory is occupied by forests.
The Altai Krai is located in the south-east of Western Siberia. The region contains almost all natural zones of Russia-the steppe and forest steppe, taiga, and mountains. The climate of the Altai Territory is highly heterogeneous because of various geographical conditions. Foothills have a temperate climate, the transition to continental.

Samples
The samples are obtained from different geographic parts (ecologically and climatically different districts) of the Tomsk region, including districts with a high beekeeping activity (the southern districts) or districts with a low apicultural activity (the northern districts), according to the local knowledge of specialists from the Society of Beekeepers. Honeybees from the apiaries of the Krasnoyarsk Krai and the Altai Krai were also investigated for comparison. Options of discoidal shift: II-negative (point F is located to the left of the perpendicular line); III-zero (point F located on a perpendicular line); IV-positive (point F is located to the right of the perpendicular line). Designation of sells: 1, radial; 2, cubital; 3, discoidal.

mtDNA analysis
DNA isolation and polymerase chain reaction (PCR) was carried out according to standard techniques with some modifications [5,6]. To amplify the COI-COII mtDNA locus, the following sequences of primers were used: 3′-CACATTTAGAAATTCCATTA, 5′-ATAAA-TATGAATCATGTGGA [5]. Amplification products were fractionated in 1.5% agarose gel, and the results were documented with the use of Gel-Doc XR+.

Microsatellite analysis
Variability of eight microsatellite loci was studied: А008 (=A8), Ap049, AC117, AC216, Ap243, H110, A024, and A113. PCR was performed using specific primers and reaction conditions according to Solignac et al. [7]. Amplification products were analyzed with ABI Prism 3730 Genetic Analyser (Applied Biosystems, Inc., Foster City, CA) and GeneMapper Software (Applied Biosystems, Inc.). Two microliters of PCR products were mixed with GeneScan500-ROX size standards (Applied Biosystems, Inc.) and deionized formamide. Samples were run according to the manufacturer's recommendations. These genetic parameters were calculated: allelic frequencies and standard error.

Infestation of honeybees by Nosema
From 10 to 70 bees were randomly selected from each bee colony and were examined for the presence of Nosema. Bee samples were stored in 70% (v/v) ethanol at room temperature prior to testing. The analysis was performed separately for each bee. The midgut of each sample was isolated, and one part of the midgut was used for the detection of Nosema spores under a light microscope, while the other part was used for DNA extraction. The midgut was suspended in 200 μL of distilled water and examined by dark-field microscopy for the presence of Nosema spores [8]. DNA was extracted from the midgut using a DNA purification kit, PureLink™ Mini (Invitrogen, Carlsbad, CA), according to the manufacturer's protocol.
In addition to the use of specific primers and fragment size to identify the species present, a selection of fragments (both N. ceranae and N. apis) was verified by DNA sequencing. Sequencing was done in both directions using forward or reverse primer (BigDye Terminator v3.1 Cycle Sequencing Kit, Applied Biosystems, Foster City, CA). DNA sequencing was performed using ABI Genetic Analyzer 3730 (Applied Biosystems) according to the manufacturer's protocol.

Results and discussion
Using the mtDNA analysis (locus COI-COII), we performed molecular genetic analysis of bee colonies (5-6 samples from each bee colony) to determine the origin of bee colony on the maternal line.
(typical for southern races). We established that 64% of bee colonies on the maternal line originate from the Middle Russian race, 28% of colonies originate from southern subspecies, and 8% are mixed bee colonies. The southern parts of the Tomsk region (with a high beekeeping activity) show a higher genetic diversity of honeybees as compared with the northern regions, which are dominated by bee colonies (96%) and apiaries (73%) that are homogeneous for the genetic variant of locus COI-COII. The bee colonies derived from the Middle Russian breed were genetically heterogeneous for the COI-COII locus: the PQQ variant was registered in 86.1% of the total number of bee colonies of the Middle Russian race, PQQQ was registered in 9.4%, and another 4.5% of bee colonies showed the presence of individuals with both allele PQQ and allele PQQQ.
Based on the analysis of mtDNA (locus COI-COII), assessment of the genetic diversity of the honeybee in apiaries of the Tomsk region has shown that the genetic structure of bee populations in the Tomsk region is complex and mosaic, especially in the southern parts of the region (Figure 3). No large areas with an array of bees having a homogeneous genetic (race) composition and maternally originating from the Middle Russian race have been found; a few apiaries were revealed, in which all bees originated from the Middle Russian breed.
In the study of variability of the COI-COII mtDNA locus in honeybees from apiaries of the Krasnoyarsk Krai and the Altai Krai, two variants of the COI-COII locus specific for Middle Russian race were identified: only variant PQQ was registered in honeybees of Krasnoyarsk Krai (Yenisei population) and two variants (PQQ and PQQQ) were found in honeybees from the Altai Krai. No a variant Q specific for southern races of bee was detected.
Due to the fact that mtDNA analysis allows assessing only the maternal component in the genome of the honeybee, bee colonies were investigated by the morphometric analysis to identify the characteristics of both the maternal and paternal lines, and to assess the level of hybridization.

Morphometric study of honeybees
The results of the morphometric study of honeybees from examined regions of Siberia (the Tomsk region, the Krasnoyarsk Krai, and the Altai Krai) were different.
According to the morphometric study, the majority of the studied bee colonies of the Tomsk region are hybrids between the Middle Russian race of bees and bees of southern origin (predominantly Carpathian race). Data on the distribution of subspecies and hybrids in the apiaries of the Tomsk region on the basis of cubital index are shown in Figure 4. Some of the apiaries, which cultivate the Middle Russian bees, were found in the northern and southern parts of the Tomsk region.  Bee colonies obtained from isolated apiaries of the Krasnoyarsk Krai are of considerable interest. The area with these isolated apiaries was not influenced by other subspecies of honeybee for many years, and all studied bees had only variant PQQ of the locus COI-COII mtDNA. However, when comparing the data of the morphometric study of bees from isolated apiaries with the Russian and European standards of the Apis. m. mellifera, the decrease of the lower limit values of cubital index was observed in the studied bees, and, as a result, for most bee colonies the deviation from the mean values of cubital index was shown ( Table 1). There are several possible explanations for the results. First, this may be the result of genetic drift, the effect of which may be because of the fact that these apiaries are isolated and there are a limited number of bees. Second, the large scale of variability of the cubital index is the result of adaptation to the environment in more severe climatic conditions. Nevertheless, these isolated apiaries in the Krasnoyarsk Krai may be considered as a unique population of the Middle Russian bee that exists for a long time without affecting other subspecies of honeybee.  The results of morphometric analysis confirmed the origin of bee colonies of Altai population from the Middle Russian race, but some influence of the southern races have been shown. For example, the parameter "Discoidal shift" deviates from the Russian breed standard: individuals with a positive value and zero of discoidal shift were found in bee colony No. 7 ( Table 2).
If bee colonies from the Krasnoyarsk Krai were obtained from the territory distant from the center and located in sparsely populated areas, in the taiga, the bee colonies from the Altai Krai inhabit the territory, characterized by high development of beekeeping and a constant active importation of bees of different origins.

The accordance of morphometric parameters and data of mtDNA analysis in honeybees in Siberia
The results of the outward morphological characters-based diagnostics of honeybees (the cubital index, the hantel index, and the discoidal shift) received from 11 bee colonies differing in the variants of the COI-COII mtDNA locus are presented ( Table 2). Only for 4 of the 11 bee colonies, a full compliance with the criteria of the breed according to the morphometric and mtDNA analysis (the three Apis mellifera mellifera colonies and one family of Apis mellifera carpatica) was shown. The remaining seven colonies are hybrid, and for three colonies a significant imbalance between genetic and morphometric parameters was shown. Hence, in order to determine the breeds in the conditions of mass bee hybridization, it is important to consider not only the features of mtDNA, but morphometric parameters as well, among which the discoidal shift is probably the most important.
These data are consistent with the results of the research of hybrid apiary, where for many years (over 30) the Middle Russian bee was bred, but the last 10 years, the southern races have been actively imported [6]. More than 50% of individuals refer to the southern races according to mtDNA analysis (variant Q of the locus COI-COII; "southern" mitotype). But none of these individuals corresponded to the southern race according to morphometric analysis ( Table 3).
In 33% of cases, individuals with "southern" mitotype had two morphometric features characteristic to the Middle Russian race.
For bees, originating from the Middle Russian race (variant PQQ of the locus COI-COII), full compliance between mitotype and morphometric parameters was found in approximately 6% of the individuals. 18% of bees had mitotype and two morphometric parameters which specific to the Middle Russian bees.
This indicates a process of cross-breeding of Middle Russian and southern races on this apiary. However, the process of "ousting of genes" is derived differently for bees of different origin: for bees of Middle Russian race the process of "ousting of genes" is smaller in scale, as among individuals with variant PQQ a smaller percentage of bees with "southern" morphometric characters was registered in comparison with the same data shown for bees with "southern" mitotypes. Lim, limits of values; М, arithmetic mean; sd, standard deviation. * Breed indicated according to the data of mtDNA analysis. ** European breed standard based on values of cubital and hantel indexes [12]. *** Russian breed standard. Discoidal shift are given according to Russian standards.  Thus, the result of study of hybrid apiaries and bee colonies indicate, on the one hand, the importance and the necessity of a comprehensive approach to the exact characterization of honeybee races. On the other hand, the results are of scientific interest for the study of genetic processes during hybridization of different subspecies of honeybee and for analyzing the process of "ousting of genes" of one race by genes of other race. For example, hybridization between the Middle Russian bee and Carpathian bee is of interest because the races belong to different evolutionary branches.
For such studies, microsatellite loci are the most informative molecular genetic markers. Microsatellite markers can be useful for the study of genetic structure of different honeybee populations and bee colonies, evaluation of genetic diversity and introgressive hybridization, differentiation of different subspecies (ecotypes), the establishment of evolutionary relationships and adaptive features of four evolutionary branches (A, M, C, and O), mapping quantitative trait loci (QTL), and search of genetic markers associated with economically significant characteristics [3,7,.
Characterization of the allele spectrum of microsatellite loci and analysis of their variability in subspecies, colonies, and individuals in the honeybee populations is the initial stage of any of the above research.

Microsatellite analysis
Variability of eight microsatellite loci (А008 (=A8), Ap049, AC117, AC216, Ap243, H110, A024, and A113) in honeybee from Siberian region was studied. Seven loci were polymorphic and only for AC216 locus one homozygous genotype was registered in all the studied bees (allele 91 bp). For each locus, the range and frequency of genotypes and alleles were determined (  Microsatellite loci differed in variability: the minimum number of alleles was detected for loci AC117 and H110 (3 alleles) and the maximum number of alleles was registered for loci A008 (10 alleles) and A113 (11 alleles). At the same time, for six of the seven polymorphic loci (except locus A024), one major allele with a frequency of more than 0.5 (from 0.5278 for allele "255"of locus Ap243 to 0.8211 for allele "183" of locus AC117) was registered regardless of the number of detected alleles.
To identify the features of honeybee from different geographical areas, the comparative analysis of the variability of the studied loci was carried out for the bees of Apis mellifera mellifera (= dark-colored forest bee, Middle Russian race) of four populations (Siberia, the Urals, and Europe) using our own data (Tomsk region and Krasnoyarsk Krai) and literature data [15,16,47] (Tables 5 and 6). The Ural population (Bashkir population) located in the nature reserve is a unique population of the dark-colored forest bee (Burzyan bee).  [47]. ** data from references [15,16]. The minimum and maximum values of allelic frequencies represented for loci A008 and A113 in honeybees of France populations; allelic frequencies for locus A024 are given for bees of only Northern France population.  Siberian populations (Tomsk region and Krasnoyarsk Krai) are closest in spectrum and allele frequencies of most studied loci (A008, Ap049, A113, Ap243, H110). The Ural population located to the west of Siberian region differs from Siberia for some loci: for locus A008 differences were registered in the spectrum of alleles, for the locus A024-in the frequency of alleles, for the loci Ap049 and Ap243-in both the spectrum and frequency of alleles. It is remarkable that the Ural population has a greater similarity in the spectrum of alleles of loci A024 and A008 to European populations.
The differences in the spectrum of alleles and the frequency of allele registration for locus A008 were revealed in honeybees of Siberia, the Ural, and European populations. For honeybees of the Ural and Europe, shorter alleles of locus A008 were predominant (154 bp and 148 bp, respectively), whereas for bees from Siberia allele "162" was the most specific. Probably this locus should be considered as a marker related to geographic and environmental conditions (specific adaptation to local conditions) [1,3,48,49] because the different populations of darkcolored forest bee (European, Ural, and Siberian populations) were compared in this study.
For some loci, for example A113, allelic spectrum overlaps, but the frequency of the alleles was different in honeybees of different populations. Different factors of population dynamics (such as founder effect, genetic drift, natural selection) can be causes of this phenomenon.
Thus, it is shown that for some loci the specific distribution of genotypes and alleles were detected in the bees, which differ by geographical location. Further research is needed and the expansion of gene-geographic studies of honeybee is relevant.
To assess the informativeness of studied loci for the differentiation of different subspecies of honeybee, the comparison of the spectrum of predominant alleles in bees of different evolutional branches (M and C) and from different geographical localization was conducted ( Table 7). Comparison of the data on the variability of microsatellite loci studied in bees of different origin and different geographical location allows making some conclusions and adjustments with respect to informativeness of these loci as markers for differentiation of subspecies of honeybee.
For locus A008, the differences in the spectrum of the most common alleles are registered between the Apis m. mellifera living in different geographical regions (as shown above), and between the two southern races (Apis m. caucasica and Apis m. carpatica).
For locus A113 clear differences in length of the most frequently detected allele were not detected both among bees of a common origin and between bees belonging to different races. Probably this locus cannot be considered informative for determining of the subspecies.
Loci A024 and Ap049 are of considerable interest for further study as candidate markers for inclusion in the diagnostic panel, differentiating subspecies. So, in general, for the locus A024 the majority of bees and bee colonies Apis m. mellifera, regardless of their habitat, are characterized by shorter length of alleles. Perhaps, for locus Ap049 the differences exist in the allelic spectrum between bees belonging to different races.  [47]. 4 Data on the European population are taken from references [15,16]. Table 7. Comparative analysis of the frequency of the most common alleles * of microsatellite loci in honeybees of different maternal origins and geographic localization.
In order to determine the subspecies status of an individual honeybee, a honeybee colony, or a honeybee population, it is important to compare allelic counts and genotypes across different studies. However, no standard reference material, such as a standard allelic ladder, is available for honeybees [3]. In addition, the spectrums of analyzed microsatellite markers often do nоt overlap and primary data on the allele spectrum and allele frequencies are not always presented in publications. In general, the present stage of the study of variability of microsatellite loci in Apis mellifera can be considered as a period of accumulation of information. At this stage of the study of honeybee it should be with caution relate to the use autosomal loci to determine the subspecies of honeybee.

Infestation of honeybees by Nosema in Siberia
Importation of races of southern origin to the territory of Siberia, where the Middle Russian breed for a long time lived, on the one hand, led to a massive hybridization of bees, a loss of purebred, decreased immunity, and increased incidence of bees. On the other hand, the import of bee families from other areas (the European part of Russia, Uzbekistan), disadvantageous in the epidemiological situation, led to the spread of diseases that have not previously registered in the territory of Siberia.
This situation was evaluated for nosemosis: the distribution Nosema infection throughout Siberia was studied, the species of microsporidia were determined, and the origin of bee colonies infected with Nosema was investigated.
Nosemosis is a parasitic disease of adult honeybees (Apis mellifera L.) caused by two described species of microsporidia, Nosema apis [50] and Nosema ceranae [51]. The disease occurs throughout the world, causes significant detriment to honey production, and results in economic losses. The original assumption was that N. apis specifically infects the European honeybee A. mellifera, causing nosemosis, and that N. ceranae is a specific pathogen of the Asian honeybee, A. cerana. Recently, it became evident that N. ceranae is also widespread in the A. mellifera population throughout the world and is already found in North and South America, across Europe and Asia [52][53][54][55][56][57][58]. It has been subsequently detected across Canada and the United States [59,60] and has been confirmed in Central America [61], Australia [62], and North Africa [63].
The geographical distribution of Nosema in Russia is not well known [64,65]. In addition, information on the prevalence of N. ceranae in Russia, including Siberia, is not complete [66]. Previously, nosemosis in honeybees in Siberia was attributed exclusively to N. apis. The problem of the distribution of Nosema and the consequences of infection for honeybees has not yet been resolved. The effects of the Nosema infection on survival and productivity of honeybees are not well studied.
For the period of 2012-2015, a screening study of 132 bee colonies from 68 apiaries of Siberia for the presence of Nosema spores was carried out [65]. For an objective evaluation, the different methods were used: microscopy and PCR. We found that honeybees of 33 colonies from 132 studied (25.0%) and 21 apiaries from 68 studied (30.9%) had spores detectable by light microscopy. As it is difficult to distinguish N. ceranae and N. apis morphologically, a PCR assay based on 16S ribosomal RNA has been used to differentiate N. apis and N. ceranae. To characterize further the identity of which species of Nosema was present, we performed PCR using primers specific for either N. apis or N. ceranae. Nosema positive samples (determined from light microscopy of spores) of adult worker bees from 33 bee colonies (21 apiaries) were tested to determine Nosema species using PCR primers of the 16S rRNA gene specific for N. ceranae or N. apis.
The samples of 28 bee colonies from 33 infected colonies (84.8%) from 19 apiaries were positive by PCR using N. apis specific primers, and the samples from three colonies (3/33, 9.1%) were positive for N. ceranae (only two of apiaries). Samples co-infected with both N. ceranae and N. apis were registered in two bee colonies (2/33, 6.0%) from two apiaries. To confirm the PCR findings, the DNA fragments were sequenced. Sequence analysis revealed a complete sequence identity for N. apis (GenBank Accession No U97150) and N. ceranae (GenBank Accession No DQ486027).
Nosema-infected bees were found in samples collected from five districts and mainly in the southern climatic areas (temperate continental parts of Siberia) (Figure 5). In the northern district ( Figure 5, C -Chainsky) bees infected by Nosema apis are imported from Uzbekistan. It was established that Nosema ceranae revealed in bees from the southern districts of the Tomsk region ( Figure 5, Shegarsky and Tomsky districts) was introduced with infected bees from southern regions of Russia.
The studied bees from apiaries of Krasnoyarsk Krai and Altai Krai were not infected with Nosema.
Reports on the impact of N. ceranae infections on honeybee health and colony survival are contradictory, and various symptoms of the disease have been described [4,48,52,[54][55][56]59,60,[67][68][69][70][71][72][73][74][75][76]. Adult bees become infected by ingesting Nosema spores, which germinate in the midgut and infect cells of the midgut epithelium. Nosema infection caused by N. apis is characterized mainly by dysentery, whereas N. ceranae is described as causing death of individuals and colonies not preceded by any visible symptoms [9,68]. Nosema apis infection is restricted to the midgut epithelium [77], whereas N. ceranae has also been detected by molecular methods in other bee tissues such as malpighian tubules and hypopharyngeal glands [78]. Perhaps, N. ceranae is the most aggressive of the two Nosema species in relation to the host and appears to be replacing N. apis in some populations of honeybees.
Currently, several reasons for the widespread presence of the parasite N. ceranae in the world and its displacement of N. apis are discussed in the literature. On the one hand, nosemosis produced by N. ceranae is considered a global problem because this parasite has wide prevalence in multiple hosts [79]. Nosema ceranae is a more aggressive parasite compared with N. apis, and consequently, it is more widespread than N. apis. On the other hand, the killing of honeybee colonies by N. ceranae could be a regional problem rather than a global phenomenon [80], and the virulence of N. ceranae could be influenced by climatic conditions [81][82][83][84][85] or might actually depend on honeybee race and honeybee genetic diversity [4,48,74,75,[86][87][88].
It is assumed that the level of infestation in honeybees can be associated with the race and the origin (local or non-local) of the bees. Some differences in the resistance to Nosema have been shown in Russian bee breeds [86]. Levels of N. ceranae infestation differed significantly between lineages and colonies for both Russian and Italian workers [87]. Unlike genetically homogeneous Italian lines [87], Russian bee lineages have a high genetic diversity and are characterized by high resistance to disease. Differences in infection levels were significant between local and introduced bee colonies [4,74]. The use of local honeybees provides a higher chance of colony survival because of their adaptation to regional environmental factors such as climate and vegetation [48,75].
To determine if the infection incidence of bees by Nosema is associated with the races of bees, we analyzed breeds of bee derived from Nosema-positive colonies using morphometric (wing venation) and molecular-genetic (mtDNA) analyses ( M ± m, average value of the sign ± the standard error of the mean. * Thirty samples of bees were examined in each family. ** Definition of subspecies was carried out based on European standard honeybee [12]. At present, the cold climate is considered as one of the limiting factors of N. ceranae distribution. It appears that the spread of N. ceranae across the globe is reduced in colder climates [81,82], as N. ceranae spores are capable of surviving high temperatures (60 °C) and desiccation, but they are intolerant of cold (4 °C) [81,82,89]. The marked decrease in N. ceranae spore germina-tion was observed after even a short exposure to low temperatures (4 °C) [82]. In warmer climates, N. ceranae is more competitive than N. apis [48,82], but the spores of N. ceranae appear to be much more vulnerable than the spores of N. apis, in particular, to freezing, and the apparent replacement of N. apis for N. ceranae remains enigmatic [83].
The different prevalence of N. ceranae may simply reflect its time of arrival, by natural spread or by the importation of infected honeybees, and mobility of bees within a country. Reduced or inhibited N. ceranae spore germination at low temperatures should hamper the infectivity and spread of this pathogen in climatic regions characterized by a rather cold winter season [82]. The presence of N. ceranae in the Tomsk region (Western Siberia, Russian) was reported previously by us [65,66] confirms the fact of a widespread N. ceranae infection in honeybee population throughout the world. However, we found N. ceranae-infected bee colonies in cold climate with long winters and humid summers, and this parasite is not associated with colony depopulation or honeybee collapse. We established that these previously infected colonies had been imported from other areas of Russia. The fact that N. ceranae is registered in the territory of Siberia with its severe climatic conditions does not agree with data on a weak survival of spores at low temperatures. At the same time, the colonies infected with Nosema (N. apis or N. ceranae) are found predominantly in the southern areas of the Tomsk region, which is characterized by more developed beekeeping and active delivery of breeds of southern origin (A. m. caucasica and A. m. carpatica) that leads to massive honeybee hybridization. Introgressive hybridization modifies the genetic pool of local honeybee populations, leading to the loss of their genetic identity [4]. The process of hybridization of different subspecies of honeybees can cause a destruction of evolutionarily developed gene complexes, leading to a decrease in the adaptive properties of organisms and populations and to a change in biological and economically significant characteristics of honeybees. The observed widespread hybridization of honeybees and the formation of hybrid bees will certainly contribute to the spread of disease.
In our research, the majority of bee colonies infected by Nosema were hybrids. This finding is consistent with the view that hybrid forms are poorly adapted to changing environmental conditions and less resistant to the disease. Therefore, our results on the Nosema infestation of bee colonies are not surprising. At the same time, it is impossible to make a conclusion about the pathogenicity of a parasite based on our data. Perhaps, hybrids are characterized by other developmental conditions of the parasite in comparison with pure breeds that do not realize the pathogenicity of N. ceranae in the host. Also, there is an open question about the distribution of a Nosema in the northern part of the Tomsk region (influence of a cold climate, insignificant number of hybrids, etc.) where the colonies infected with Nosema were not detected except Chainsky district (N. apis-infected bees were imported from Uzbekistan). Siberia can be an ideal location to study how the spread of this disease correlates with climatic conditions and how the disease moves to particularly remote areas. This is an especially intriguing thought since changes in disease prevalence and pathogen virulence because of climatic change are widely discussed [80]. Obviously, more research is needed to elucidate the full effect of N. ceranae infection in A. mellifera colonies in different geographical areas and to understand if individual virulence levels and colony virulence levels differ between the two parasites.

Conclusion
This study of honeybees in Siberia shows the need for a comprehensive approach to the study of various aspects of the honeybee, such as differentiation of subspecies, the role of environmental (geographical) factors in the formation of the genetic diversity of bees, and the incidence of bees.
The primary task of the study of the genetic diversity of honeybees is to determine their subspecies composition. When performing gene-geographical research, it is important to consider the assessment of adaptive and selective significance of genetic markers. This is also important for the planning and conducting of works having applied nature.
Along with exterior characters used for a long time to identify the breed of honeybees, molecular genetic techniques are actively applied. However, in connection with the high level of hybridization of bees, when about one-third of bee colonies show an imbalance between genetic and morphometric parameters, and in some cases, their complete mismatch occurs, a comprehensive analysis of the bees is necessary.
The presence of hybrid forms in an area where the genetic diversity is studied, on the one hand, creates unfavorable background for conservation of gene pools of unique subspecies (for example, dark-colored forest bee), on the other hand, makes it difficult to search for adaptively significant and economically valuable traits (possible distortion of results and their interpretation). Therefore, it should be taken into account in conducting such studies. The above data also indicate that only the exterior or just genetic traits may be insufficient to determine the origin of bees and only the simultaneous analysis of morphometric parameters and data on the variability of locus COI-COII of mtDNA allow to evaluate the breed and cases of hybridization objectively.
In the conditions of widespread crossbreeding of bees, genetic methods to control the purity of bee colonies must also be improved. Research in this direction is carried out by international and Russian researchers [43,47,90]. Therefore, on the basis of extensive research carried out on the territory of Eastern Europe (search of informative markers was conducted among more than 1,000 SNP using five different analytical methods), five panels, consisting of 48,96,144,192, and 284 markers informative for determining the ancestral origin of species have been developed. The authors propose to use the results of this study to identify and evaluate the impurity of C-lines (in particular, Apis m. ligustica and Apis m. carnica) to the M-line (Apis m. mellifera) [43]. Russian researchers have only begun such studies, but the results obtained at this stage suggest that populations of honeybees living on the territory of Russia are characterized by wide genetic variability, and it is unlikely to develop a uniform panel of markers for the entire territory of the Russia for differentiation of the various breeds of bees. It is necessary to integrate the scientific achievements and results of the various laboratories and scientific groups of all over the world to establish general regularities of the genetic variability of the bees and to assess the adaptive and selective potential of honeybees.