Molecular identification and phylogenetic analysis of marmots in the Xinjiang's Altai Mountains, a newly discovered plague focus of China

TL;DR

Scientists confirmed marmots in China's Altai Mountains are Marmota baibacina using DNA, aiding plague control efforts.

Contribution

First molecular confirmation of marmot species in Xinjiang's Altai Mountains and identification of a new plague focus.

Findings

Marmots in the Altai Mountains belong to Marmota baibacina based on COI, Cytb, and D-loop gene analysis.

Altai and northern Tianshan marmots are closely related genetically, suggesting a shared subspecies.

The Altai plague focus is classified as M. baibacina–Spermophilus undulatus due to both being reservoir hosts.

Abstract

The Altai Mountains in Xinjiang have been recognized as a new plague focus in China since 2023. Although previous morphological studies suggested that marmots in this region belong to Marmota baibacina, this classification remained unconfirmed at the molecular level. In this study, 37 marmot samples were collected: including 10 M. baibacina from the northern Tianshan Mountains, 9 M. baibacina from the southern Tianshan Mountains, and 18 unverified individuals from the Altai Mountains. The COI, Cytb, and D-loop gene fragments were sequenced and subjected to phylogenetic analysis. Homology comparisons revealed greater than 95% sequence identity across all samples, with all matches corresponding to marmot populations. Genetic distances between groups (1%–7%) exceeded those within groups (0%–2%). Fst values among populations ranged from 0.55780 to 0.98449, indicating significant differentiation, particularly between the southern Tianshan population and the other two groups. In contrast, genetic divergence between the Altai and northern Tianshan populations was minimal. Maximum likelihood trees based on COI and Cytb sequences strongly supported a close phylogenetic relationship between the northern Tianshan and Altai marmots, which clustered together, while the southern Tianshan population formed a separate clade. The D-loop region provided lower phylogenetic resolution. Based on these results, we confirm that the marmots in the Altai Mountains belong to M. baibacina and classify this plague focus as a M. baibacina–Spermophilus undulatus focus, given that S. undulatus is also a documented reservoir host in the area. These findings provide a molecular basis for targeted plague surveillance and control. The close genetic affinity between the northern Tianshan and Altai populations further suggests they may represent the same subspecies.

•The Xinjiang's Altai Mountains have been identified as a new plague focus in China since 2023.•First study to confirm the species of marmots distributed in this region at the molecular level.•The marmots distributed in this region belonged to the species Marmota baibacina.•The plague focus in this region is classified as a M. baibacina–Spermophilus undulatus focus.•The northern Tianshan and Altai marmots may represent the same subspecies.

The Xinjiang's Altai Mountains have been identified as a new plague focus in China since 2023.

First study to confirm the species of marmots distributed in this region at the molecular level.

The marmots distributed in this region belonged to the species Marmota baibacina.

The plague focus in this region is classified as a M. baibacina–Spermophilus undulatus focus.

The northern Tianshan and Altai marmots may represent the same subspecies.

Figures

Phylogeny of *M. baibacina* from different regions in Xinjiang based on *COI*. A: Geographic distribution of marmot sampling sites in Xinjiang, China. B: ML tree constructed based on the *COI*.Fig. 1Fig. 2Phylogeny of *M. baibacina* from different regions in Xinjiang based on the *Cytb*. A: Geographic distribution of marmot sampling sites in Xinjiang, China. B: ML tree constructed based on the *Cytb*.Fig. 2Fig. 3ML tree constructed based on the *D-loop*.Fig. 3

Full text

Introduction

1

The Altai Mountains span China, Mongolia, Russia, and Kazakhstan, with the Chinese section located in northern Xinjiang [1]. Plague foci are distributed across all four countries within the Altai Mountains range [[2], [3], [4]]. The western extent of the Mongolian Marmota sibirica plague focus reaches the northern Baitag Bogd region along the China-Mongolia border [5]. However, it remains unclear whether M. sibirica is present within the Chinese Altai Mountains. Since 2020, plague F1 antigen and anti-F1 antibodies have been repeatedly detected in marmots and Spermophilus undulatus in this region, and Yersinia pestis has been isolated [6,7]. Consequently, the Altai Mountains in Xinjiang have been identified as a new plague focus in China. Nevertheless, the local marmot species has not been definitively identified. Previous morphological studies suggested that the marmots distributed in the Altai Mountains of Xinjiang belong to M. baibacina, but this classification lacks molecular confirmation [8]. Therefore, accurate identification of the reservoir host in this newly discovered plague focus is essential for implementing targeted prevention and control strategies.

DNA barcoding, a molecular identification method first proposed by Hebert et al. (2003), employs a standardized short genomic fragment to facilitate rapid and cost-effective species classification [9,10]. In animal studies, the most widely adopted barcoding markers are derived from mitochondrial DNA (mtDNA), particularly the cytochrome c oxidase subunit I (COI) and cytochrome b (Cytb) genes. These loci possess several characteristics ideal for barcoding, including compact size, structural simplicity, high amplification success, and considerable interspecific sequence divergence [11]. Guo et al. applied this approach to identify and classify rodents, including Marmota himalayana, in plague foci in Gansu Province, with satisfactory results [12]. Despite this successful application, studies specifically employing DNA barcoding within the genus Marmota remain scarce, highlighting a gap that warrants further systematic investigation.

To address this gap, this study aimed to evaluate the efficacy of DNA barcoding for species delineation within Marmota and to elucidate the phylogenetic relationships of M. baibacina across Xinjiang, thereby providing a molecular basis for characterizing the plague focus in the Altai Mountains of China. We collected M. baibacina specimens from the northern and southern Tianshan Mountains alongside marmot samples from the Altai Mountains. Three mitochondrial genes (COI, Cytb, and D-loop) were selected as candidate barcodes for subsequent molecular identification and phylogenetic analysis.

Material and methods

2

Study area

2.1

The study area encompassed the Altai Mountains and the northern and southern Tianshan Mountains in Xinjiang. The Altai Mountains, characterized by a diverse landscape of alpine meadows, taiga forests, and high-altitude steppes, experience a continental climate of harsh, cold winters and mild summers. The northern Tianshan Mountains feature mid-altitude coniferous forests (e.g., Picea schrenkiana) and montane grasslands under a semi-humid climate, marked by higher windward precipitation (400–800 mm annually) and pronounced seasonal temperature shifts. Conversely, the southern Tianshan Mountains are distinguished by arid to semi-arid conditions, supporting only sparse xerophytic shrubs and alpine deserts, alongside low precipitation (<300 mm annually) and extreme diurnal temperature fluctuations. A total of 6 sampling sites were selected (Fig. 1): three in the Altai Mountains (Qinghe County, Fuhai County and Altai City), two in the northern Tianshan Mountains (Urumqi County and Wusu City), and one in the southern Tianshan Mountains (Wuqia County).

Sample collection and processing

2.2

All animal procedures were approved by the Institutional Review Board of Xinjiang Center for Disease Control and Prevention (protocol No.XJCDC2022-14) and conducted in strict accordance with the guidelines of the Institutional Animal Care and Use Committee (IACUC) as well as relevant national regulations.

In this study, 10 M. baibacina individuals were collected from the northern Tianshan Mountains (Urumqi County: WX01∼05, and Wusu City: WS01∼05), 9 from the southern Tianshan Mountains (Wuqia County: WQ01∼09), and 18 unverified marmots were collected from the Altai Mountains (Qinghe County: QH01∼06, Fuhai County: FH01∼05, and Altai City: ALT01∼07). Among them, QH03, QH05 and QH06 were decomposed marmot carcasses (Table 1). All marmots were collected from typical burrowing habitats within known or suspected plague-active zones. Following necessary safety protocols, marmots were euthanized via intraperitoneal injection of sodium pentobarbital. Liver tissue was then aseptically collected and immediately stored at −80 °C for subsequent analysis. Live was selected as the primary source for DNA extraction due to its high genomic DNA yield and stability in decomposed carcasses, a common condition in field surveillance, where blood samples are frequently unavailable or degraded.Table 1. Characteristics of marmot sampling locations.Table 1. LocationsSampling dateNumber of marmots collectedGPS coordinatesElevationHabitat typeUrumqi CountyJune 15, 2024543.3447°, 87.0541°2573 malpine meadowWusu CityJune 23, 2024543.5903°, 85.0234°1882 malpine meadowWuqia CountyJuly 10, 2024940.4133°, 75.2611°3439 malpine meadowQinghe CountyJuly 19, 2024646.6460°, 90.7621°2752 malpine meadowFuhai CountyAugust 5, 2024542.5648°, 87.3917°2236 malpine meadowAltai CityAugust 12, 2024747.7888°, 87.8414°1875 malpine meadow

DNA extraction

2.3

Genomic DNA was extracted from liver tissue using an Animal Tissues/Cells Genomic DNA Extraction Kit (Solarbio Science & Technology Co., Ltd., Beijing, China). Briefly, tissue samples were ground in liquid nitrogen. The extraction procedure was conducted in accordance with the manufacturer's instructions, supplemented with an RNase A treatment to remove residual RNA. DNA purity was assessed by measuring the A260/A280 ratio (>1.8) using a Nanodrop spectrophotometer, and integrity was confirmed by 1% agarose gel electrophoresis. Purified DNA was stored at −20 °C until further use.

Primer design

2.4

The primers for amplifying the COI, Cytb, and D-loop gene fragments were designed using Primer 5.0 software, based on publicly available mitochondrial sequences of the four marmot species found in China: Marmota caudata (GenBank: JF499307.1 and AF143924.1), M. himalayana (GenBank: JX962054.1, OR389671.1, and NC_018367.1), M. baibacina (GenBank: KX859255.1, MT412466.1, and NC_086592.1) and M. sibirica (GenBank: KX859268.1, MT412467.1, and PP357272.1). These primers did not form dimers or hairpin structures, and were synthesized by Xinjiang Ouyi Biotechnology Co., Ltd. The primer sequences were shown in Table 2.Table 2. Primer sequences.Table 2. Amplified fragmentPrimer nameSequence(5′ to 3′)COⅠ**COⅠ-FCGGTATAGTAGGAACTGCRCTCACOⅠ-RCCAGCAGGATCRAAAAATGTAGTCytb**Cytb-FGCAACCGTAATYACYAATCTCCytb-RGGTTGYCCTCCRATTCAGD-loop**Dloop-FCATGCATATCAAGCACGTTCATAATACDloop-RTTAAGCCCGACTATGGCAGAT

PCR amplification

2.5

PCR amplifications were performed in a 50 μL reaction volume containing 25 μL of 2 × Super Taq PCR Mix (Wuhan Genenode Biotech Co., Ltd.), 1 μL of each forward and reverse primers (10 μM), 2 μL of template DNA, and 21 μL of nuclease-free water. The thermal cycling protocol consisted of an initial denaturation at 94 °C for 3 min; followed by 35 cycles of 94 °C for 30 s, 57 °C for 30 s, and 72 °C for 36 s; with a final extension at 72 °C for 5 min. PCR products of the expected size, as verified by agarose gel electrophoresis, were purified using an E.Z.N.A.® Gel Extraction Kit (Omega Bio-tek, Inc., Norcross, GA, USA). Purified amplicons were then submitted to Xinjiang Ouyi Biotechnology Co., Ltd. for bidirectional sequencing.

Data analysis

2.6

The SeqMan and EditSeq program within DNAStar software package were used to assemble the sequencing results and to obtain the complete COI, Cytb and D-loop gene sequences. BLASTn searches were performed against the GenBank database. Sequences with >95% similarity and consistent morphological identification were selected as references for species verification of marmots from the Altai Mountains. Multiple sequence alignment was conducted using the ClustalW function in MEGA X, and the base composition, base substitutions, transitions, transversions, and their ratio (R) were calculated. Genetic distances within and between populations were estimated based on the Kimura 2-parameter (K2P) model. Additionally, DnaSP 6.0 software was used to calculate the fixation index (Fst) to assess intra- and interspecific genetic differentiation. Five marmot species were included in the analysis: M. vancouverensis (NC_048490.1), M. flaviventris (NC_042243.1), M. sibirica (NC_086593.1), and M. olympus (OQ694761.1) [[13], [14], [15]], with Rhombomys opimus (MK359635.1) serving as the outgroup. The maximum likelihood (ML) method was employed to construct the phylogenetic tree. First, JModeltest v2.1.7 [16] was used to determine the optimal nucleotide substitution model for ML tree construction, and the GTR + G + I (GTR + GAMMA + I) model was selected based on the Akaike information criterion (AIC). Subsequently, the ML tree was constructed using RAxML v8.2.12 [17] with 1000 bootstrap replicates. Finally, visualization and optimization of the phylogenetic tree were performed using FigTree v1.4.4 (http://tree.bio.ed.ac.uk/software/figtree/) and the Interaction Tree of Life (iTOL) website (http://itol.embl.de/) to improve clarity and presentation quality. SPSS 23.0 software was used to compare differences in nucleotide composition among marmot populations, utilizing the Mann-Whitney U test or Student's t-test. Statistical significance was set at a two-tailed P < 0.05.

Results

3

Homology comparison

3.1

BLASTn comparisons of the COI, Cytb, and D-loop sequences against the NCBI database revealed 97–100% homology with reported marmot sequences, all with >95% query coverage. All matched sequences were of Marmota origin, confirming the accuracy of PCR amplification and sequencing for downstream analyses (Table 3).Table 3. Alignment results of BLAST for the COI, Cytb and D-loop gene fragments.Table 3. Marmot populationQuery coverIdentAccessionDescriptionCOⅠNorth Tianshan M. baibacina99%99.12%KX859255.1Partial CDSSouth Tianshan M. baibacina99%99.65%KX859255.1Partial CDSAltai M. baibacina100%99.65%KX859255.1Partial CDSCytbNorth Tianshan M. baibacina97%95.33%MT412466.1Complete CDSSouth Tianshan M. baibacina98%99.33%MT412466.1Complete CDSAltai M. baibacina97%99.50%MT412466.1Complete CDSD-loopNorth Tianshan M. baibacina97%97.64%MT412446.1Partial SequenceSouth Tianshan M. baibacina98%99.46%MT412446.1Partial SequenceAltai M. baibacina100%99.11%MT412446.1Partial Sequence

Nucleotide composition and substitution statistics

3.2

Analysis of nucleotide composition across the three populations revealed that thymine (T) was the most abundant base, whereas guanine (G) was the least abundant. The A + T content was significantly higher than the G + C content (t = 9.619, P < 0.01), consistent with the typical A + T bias in vertebrate mitochondrial genomes. Regarding nucleotide substitutions, transitions exceeded transversions (R > 1) in most comparisons. Exceptions included the COI gene in the northern Tianshan M. baibacina population (R < 1), the Cytb gene in the Altai population (R = 1), and the D-loop gene in the northern Tianshan population (R = 1) (Table 4).Table 4. Results of base composition analysis.Table 4. Marmot populationBsse composition(%)A + T(%)C + G(%)iisisvRATGCCOⅠNorth Tianshan M. baibacina26.532.915.924.759.440.6571010South Tianshan M. baibacina26.732.615.725.359.341.057020-Altai M. baibacina26.632.715.824.559.340.357020-CytbNorth Tianshan M. baibacina28.832.111.827.360.939.2601431.3South Tianshan M. baibacina28.933.811.325.962.737.2603321.5Altai M. baibacina28.032.312.327.460.339.7600441.0D-loopNorth Tianshan M. baibacina25.929.915.528.755.844.2558331.0South Tianshan M. baibacina25.729.815.828.755.544.5559321.5Altai M. baibacina25.330.215.928.655.544.5551832.7

Genetic distance

3.3

The K2P genetic distances between populations ranged from 1% to 7%, with the largest value observed for Cytb and the smallest for COI. Substantial genetic divergence was detected between the southern Tianshan population and both the northern Tianshan and Altai populations, whereas the distance between the latter two was relatively low. Intraspecific distances varied from 0% to 2%, consistently lower than interspecific distances (1%-7%). Population differentiation, as measured by Fst, ranged from 0.55780 to 0.98449. Fst values between the southern Tianshan population and the other two populations all exceeded 0.75, indicating pronounced genetic differentiation. The lowest Fst values were consistently observed between the northern Tianshan and Altai populations across all genes, suggesting limited divergence. Within-group Fst values ranged from 0.71549 to 0.93946, further confirming marked genetic structuring (Table 5).Table 5. Genetic distance analysis.Table 5. Marmot populationNorth Tianshan M. baibacinaSouth Tianshan M. baibacinaAltai M. baibacinaWithin-populationCOⅠNorth Tianshan M. baibacina0.98449b0.809630.93797South Tianshan M. baibacina0.03a0.945150.93946Altai M. baibacina0.010.020.93097Within-population0.000.000.00CytbNorth Tianshan M. baibacina0.903720.749640.85759South Tianshan M. baibacina0.070.887190.86054Altai M. baibacina0.030.060.85446Within-population0.010.010.01D-loopNorth Tianshan M. baibacina0.853530.557800.73454South Tianshan M. baibacina0.040.755960.73626Altai M. baibacina0.030.040.71594Within-population0.010.010.02aThe K2P genetic distance was below the diagonal, andbThe Fst value was above the diagonal.

Phylogenetic tree

3.4

ML trees were reconstructed for each gene fragment. Phylogenies based on COI and Cytb placed the northern Tianshan M. baibacina and Altai populations within the same clade (ML bootstrap support 96–99%), indicating they belong to the same subspecies. The southern Tianshan M. baibacina formed a sister group to this clade. Outgroup taxa were clearly separated. In contrast, the D-loop-based tree showed poor resolution, with mixing between the studied marmots and outgroup taxa (Fig. 1, Fig. 2, Fig. 3).Fig. 1. Phylogeny of M. baibacina from different regions in Xinjiang based on COI. A: Geographic distribution of marmot sampling sites in Xinjiang, China. B: ML tree constructed based on the COI.Fig. 1. Fig. 2Phylogeny of M. baibacina from different regions in Xinjiang based on the Cytb. A: Geographic distribution of marmot sampling sites in Xinjiang, China. B: ML tree constructed based on the Cytb.Fig. 2. Fig. 3ML tree constructed based on the D-loop.Fig. 3

Discussion

4

Marmots (genus Marmota, family Sciuridae, order Rodentia) are large, terrestrial, burrowing rodents that serve as major reservoir hosts of Y. pestis, the causative agent of plague [18]. Their broad distribution and high natural infection rates contribute significantly to the maintenance and transmission of plague in endemic regions [19,20]. Among the four marmot species present in China, three occur in Xinjiang: M. himalayana, M. caudata, and M. baibacina [21]. M. baibacina comprises two subspecies, M. baibacina centralis Thomas 1909 and M. baibacina baibacina Brandt 1843, which are morphologically similar and challenging to distinguish based on size, pelage characteristics, or cranial characteristics alone [22].

In this study, we performed molecular identification and phylogenetic analyses of marmot specimens collected from the northern Tianshan, southern Tianshan, and Altai Mountains, utilizing three mitochondrial markers: COI, Cytb, and D-loop genes. The COI and Cytb genes were selected as standard DNA barcodes due to their high interspecific discriminatory power and the availability of comprehensive reference databases, whereas the D-loop gene was incorporated to evaluate intraspecific population-level variation. Nuclear markers were deliberately excluded, as our primary objective was rapid species identification for plague surveillance—a context in which mtDNA markers have demonstrated proven efficacy for marmot phylogenetic reconstruction in previous studies. All analyzed sequences exhibited high levels of sequence homology. Consistent with typical vertebrate mitochondrial genomic architecture, the A + T content markedly exceeded the G + C content across all loci. Genetic distance and Fst analyses revealed minimal differentiation between the Altai and northern Tianshan populations, whereas both showed significant divergence from the southern Tianshan population. Phylogenetic trees reconstructed from COI and Cytb sequences consistently indicated that marmots from all three regions belong to M. baibacina, with the northern Tianshan and Altai populations forming a sister clade to the southern Tianshan group-consistent with prior morphological identifications. Node support values for these topological relationships were moderate to high (68–99% ML bootstrap). Although the close relationship between the northern Tianshan and Altai populations suggests possible subspecific conspecificity, further investigation is required to confirm subspecific classification. Between COI and Cytb, the latter afforded stronger support for marmot monophyly, aligning with findings from Steppan et al. [23] In contrast, the D-loop gene generated poorly resolved phylogenies, including instances of outgroup mixing, indicating its limited utility for reliable molecular identification of marmots in this study.

These results conclusively identify the marmots in the Altai Mountains as M. baibacina, enabling the characterization of the plague focus in this region as a M. baibacina–S. undulatus focus. Although our sample size from the Altai region (n = 18) is limited, specimens were collected across three counties (Qinghe, Fuhai and Altai City) covering a range of elevations, which enhances geographic representation. This clarification is critical for understanding the ecological and epidemiological structure of this newly identified plague focus and facilitates the design of targeted surveillance and control strategies. Furthermore, the close phylogenetic relationship between the northern Tianshan and Altai populations suggests possible subspecific affinity, highlighting the need for additional genomic studies. It is also noteworthy that this study successfully applied DNA barcoding to identify decomposed marmot carcasses, which are frequently encountered in plague surveillance but are unsuitable for morphological identification. Thus, the approach provides a valuable molecular tool for monitoring wildlife plague hosts in the field.

The main limitations of this study were its relatively small sample size and incomplete mitogenome coverage. To fully elucidate the population genetics and evolutionary relationships of marmots in Xinjiang, future studies should expand the sampling scope to include broader geographic coverage and additional key species (e.g., M. himalayana and M. caudata), and employ complete mitogenome sequencing to further evaluate the efficacy of standard barcode genes.

Conclusion

5

In summary, the mitochondrial COI and Cytb genes are effective for species identification and phylogenetic analysis in marmots and remain applicable to degraded samples from carcasses, whereas the D-loop gene appears less suitable in this context. Phylogenetic reconstruction based on COI and Cytb sequences confirmed that the marmots in the Altai Mountains belong to M. baibacina, supporting the classification of the plague focus in this region as a M. baibacina–S. undulatus focus. Furthermore, the close phylogenetic relationship between the northern Tianshan and Altai populations suggests that they may represent the same subspecies.

Authors' contributions

Conceptualization, W.M. and X.W.; methodology, Y.Z. and J.Z.; software, Y.Z. and J.Z.; validation, R.T. and M.X.; formal analysis, Q.W., M.B. and Y.L.; investigation, A.R. and X.W.; resources, W.M. and X.W.; data curation, Y.L.; writing—original draft preparation, Q.W., M.B. and Y.L.; writing—review and editing, W.M. and X.W.; visualization, W.M.; supervision, X.W.; project administration, M.B.; funding acquisition, Q.W. All authors have read and agreed to the published version of the manuscript.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.