The Caspian barbastelle, Barbastella caspica, has spread widely in the Caspian region, Iran, and Central Asia; however, there is no evidence of its occurrence in China so far. During a field investigation, we collected a single specimen of B. caspica in China’s Xinjiang Uyghur Autonomous Region. At the same time, we obtained the free-flight echolocation calls of the bat. It omitted signals with start frequency of 33.15 ± 1.43 kHz, end frequency of 29.82 ± 0.40 kHz, frequency of most energy 31.48 ± 0.40 kHz, duration of 2.43 ± 0.24 ms, and a pulse interval of 246.57 ± 9.48 ms, which are probably type-I sounds emitted through the mouth. We also sequenced its entire mitochondrial genome to elucidate the genomic structure and its evolutionary relationships with closely related Barbastella. The mitochondrial genome of B. caspica spans 16,933 bp, comprising 13 protein-encoding genes, 22 transfer RNA genes, two ribosomal RNA genes, and a displacement loop/control region. Consistent with previous bat mitogenome reports, the majority of mitochondrial genes are encoded on the heavy chain. A phylogenetic analysis based on 13 protein-coding genes revealed that Rhogeessa, Plecotus, and B. caspica formed a clade within Vespertilionidae. Barbastella caspica was found to be a sister species to B. beijingensis and B. leucomelas in phylogenetic trees using the cytochrome b and ND1 gene sequences. This is the first report of the mitogenome of a member of the genus Barbastella, as well as the first record of the distribution of B. caspica in China and first documentation of its echolocation calls.

Echolocation calls, phylogenetic analysis, Xinjiang

The Barbastella genus is widely distributed from Northeast Africa to across Eurasia to Taiwan and Japan. Currently, only six species are recognized: B. barbastellus Schreber, 1774, B. beijingensis Zhang et al., 2007, B. caspica Satunin, 1908, B. darjelingensis Hodogson, 1855, B. leucomelas Cretzschmar, 1826, and B. pacifica Kruskop et al., 2019 (https://www.checklistbank.org/). In China, distributional records exist only for B. beijingensis and B. darjelingensis (http://www.sp2000.org.cn/).

The Caspian barbastelle, B. caspica, primarily inhabits drier habitats and is occasionally found in caves, crevices, and mines. Its main distribution encompasses northern Iran, the Caucasus region (Armenia, Azerbaijan, and Dagestan in Russia), Uzbekistan, and Tajikistan (Kruskop 2015). Research on this species is relatively limited, with a few studies focusing on taxonomic status and distribution (Kruskop et al. 2019). Furthermore, genomic studies on species of Barbastella have been lacking, and the phylogenetic position of this genus within the family Vespertilionidae has not been explored.

In this study, we used mist nets to capture and ultrasound recording equipment to record B. caspica echolocation calls. Furthermore, we conducted a comprehensive assembly and analysis of the complete mitochondrial genome of B. caspica, thus establishing the first genomic resource of Barbastella. Specifically, we analyzed the nucleotide composition of the entire mitochondrial DNA molecule, investigated the codon usage patterns and selective constraints of protein-coding genes (PCGs), and described the secondary structure of each identified tRNA gene. Finally, based on mitochondrial PCGs, cytochrome b (Cytb), and ND1 sequences, we examined the phylogenetic position of Barbastella among other representative species of Vespertilionidae and of B. caspica within its genus. On the one hand, the complete assembly of mitochondrial genome markers was a significant step toward advancing our understanding of the genomic evolutionary biology and systematics of Barbastella species. On the other hand, this study also reported the first documentation of this species in China and the features of echolocation calls during flight.

A bat individual was captured using mist nets during a survey of chiropteran resources in Yarkand County (37°54'24.75"N, 76°47'2.86"E), Xinjiang Uygur Autonomous Region of China, in July 2023 (Fig. 1). The specimen (SC230705005) is currently stored at the College of Life Sciences, Shihezi University. Morphological identification revealed that the bat had short, wide ears with the front ends of both ears connected, indicating that it belongs to a species of barbastelles bat, Barbastella genus (https://www.checklistbank.org/). A Song Meter SM4BAT FS ultrasonic recording device (Wildlife Acoustics, USA) was placed next to the mist net to record bat echolocation calls. Subsequently, the recorded echolocation sound waves were analyzed using sound analysis software (Kaleidoscope v. 5.4.8).

Figure 1. 

Map of Central Asia showing the geographic range of Barbastella caspica (green) and sampling site (red triangle) in southern Xinjiang, China.

In the laboratory, total genomic DNAs were extracted from muscle tissues using the Tiancheng Genomic DNA Extraction Kit (Tiangen Biotech, Beijing, China). The mitochondrial genome of B. caspica was amplified using PCR with 11 pairs of custom-designed primers (Suppl. material 1). Products that met quality-control criteria were purified and commercially sequenced. Sequencing data were processed and assembled using SeqMan software (Tamura et al. 2013). The annotation of the mitochondrial genome was performed using the GeSeq organelle genome annotation server (Tillich et al. 2017) (https://chlorobox.mpimp-golm.mpg.de/geseq.html). Annotation refinement and adjustment of start/stop codons were performed using MEGA X (Kumar et al. 2018). The finalized mitochondrial sequence has been deposited in NCBI GenBank under accession number PP963575.

The echolocation call of Barbastella caspica is characterized by frequency modulation (FM) (Fig. 2a, b). In free-flight outdoor conditions, the pulses are composed of a single harmonic. The peak frequency is notably low, with the highest energy peak occurring at 31.48 ± 0.40 kHz (Fig. 2c). The frequency bandwidth is narrow, measuring only 5.79 ± 1.04 kHz. The initial frequency is at 33.15 ± 1.43 kHz and the final frequency is at 29.82 ± 0.40 kHz. The pulse duration is relatively short, approximately 2.43 ± 0.24 ms, with an interpulse interval of 246.57 ± 9.48 ms (Table 1). These characteristics closely resemble the sound waves emitted by other species of Barbastella while foraging (Zhang et al. 2007) and were similar to the type-I sounds of Barbastella species (Denzinger et al. 2001). However, considering that certain species of horseshoe bat consistently emit two different types of sound waves during foraging (Seibert et al. 2015), it is possible that our sound-wave detector failed to capture type-II sounds. Alternatively, it is likely that B. caspica does not produce this particular sound during foraging or that the frequency of the emitted sound waves is lower than in other Barbastella species. These possibilities should be confirmed in future studies.

Figure 2. 

Echolocation calls features of B. caspica in free flight conditions: the spectrogram and waveform with time unit in milliseconds (a), the spectrogram with a time unit of seconds (b) and the energy spectrum (c).

The mitochondrial genome of B. caspica is a circular DNA molecule with a length of 16,933 base pairs (Fig. 3). The genome encompasses a total of 37 genes, consisting of 13 PCGs, 22 transfer RNA genes (tRNAs), two ribosomal RNA genes (rRNAs), and one D-loop region. The size and organization of these mitochondrial genes (Table 2) are consistent with previous reports of other vespertilionid species (Guo et al. 2021; Martínez-Cárdenas et al. 2024; Valencia M. et al. 2024). Among the 13 PCGs (11,408 bp), they exhibit similarities with other species of Vespertilionidae, such as being located on the heavy strand except for ND6 (Martínez-Cárdenas et al. 2024; Valencia M. et al. 2024). The average A+T content of PCGs in mitochondria is 59.92%, ranging from 56.31% (COX1) to 64.73% (ATP8), which is higher than the G+C content (40.08%) of the 13 PCGs. Furthermore, they show similar negative AT skew and CG skew, as well as a high A+G content (60.03%) (Suppl. material 2) (Guo et al. 2021; Martínez-Cárdenas et al. 2024; Valencia M. et al. 2024). All PCGs start with ATG or ATA codons and terminate with TAA or truncated T residues, except for the Cytb gene, which terminates with AGA (Table 2).

Figure 3. 

Mitochondrial genome map of B. caspica. The mitochondrial DNA of B. caspica is 16,933 base pairs long, consisting of different segments: 22 blue segments representing tRNA coding regions, 2 red segments corresponding to 12SrRNA and 16SrRNA, 7 yellow segments for ND1, ND2, ND3, ND4L, ND4, ND5, and ND6, 3 pink segments for COX1, COX2, and COX3, 1 purple segment for the Cytb gene, and 1 light red segment for the D-loop region.

Suppl. material 3 shows the codon counts and RSCU values of B. caspica. The 33 codons are used more frequently (RSCU > 1, Suppl. material 4). The codons AAU-Asn (158), ACA-Thr (133), CCA-Pro (130), ACU-Thr (121), and CUA-Leu (118) are the most frequently used. There are 22 typical tRNA genes, ranging in length from 59 bp (tRNA-Ser1) to 75 bp (tRNA-Leu2). Eight of these genes are located on the L strand, while 14 are on the H strand. In total, they span 1520 bp. Except for tRNA-Ser (Table 2, Suppl. material 5), all these tRNA molecules have the classical cloverleaf structure. This phenomenon has been mentioned in previous studies and is common among metazoans (Vivas-Toro et al. 2021; Basaldúa et al. 2023). However, more research is needed to determine the functionality of these features in B. caspica.

Based on 13 PCG sequences, we successfully constructed the phylogenetic topology of 31 species from the vespertilionid subfamilies Myotinae and Vespertilioninae. Consequently, Rhogeessa, Plecotus, Pipistrellus, Glischropus, Hypsugo, and Barbastella formed the subfamily Vespertilioninae, with Barbastella being a sister genus to Plecotus (Fig. 4). Consistent with previous results based on the COI gene (Chakravarty et al. 2020), we found that Plecotus and Barbastella belonged to the same tribe, Plecotini, which also includes four other genera (Wilson and Mittermeier 2019; https://www.checklistbank.org/), implying that more genome-based phylogeny is required to understand the intergeneric evolutionary relationships within the Plecotini.

Figure 4. 

The phylogenetic relationships of the Vespertilionidae based on 13 protein-coding genes using the maximum-likelihood method with 1,000 bootstrap replicates. Tadarida latouchei and Miniopterus fuliginosus were designated as outgroups. Nodes with support values ≥ 80 are indicated.

Phylogenetic trees were constructed to elucidate the evolutionary relationship of B. caspica with other species of vespertilionids, based on the Cytb and ND1 genes along with all PCGs. Within Barbastella, B. caspica is identified as a distinct species. However, differential topological structures were observed in the phylogenetic trees constructed based on the ND1 and Cytb genes (Fig. 5). Namely, B. caspica was a sister species to B. leucomelas in the ND1 phylogenetic tree, but sister to B. beijingensis in the Cytb phylogenetic tree.

Figure 5. 

The Bayesian analyses of phylogenetic relationships of members of the Barbastella genus based on 806 bp ND1 (left) and 1140 bp Cytb (right) sequences using Bayesian-inference (BI) and maximum-likelihood (ML) methods. Rhogeessa mira is used as the root, and nodes with support values of ≥ 0.7 (BI) and 80 (ML) are labeled.

The pairwise distances (Table 3) shows that the smallest genetic distances (3.7% based on the ND1 gene) are between B. caspica and B. leucomelas. However, the direct pairwise distance between the two species based on the Cytb gene is 13.4%, which is consistent with the results of the phylogenetic tree.

The systematic construction of the ND1 phylogenetic tree, as well as the ND1 genetic distances within Barbastella, consistently indicate a close genetic relationship between B. caspica and B. leucomelas, which agrees with the results of Smirnov et al. (2020). In contrast, the findings of the Cytb analyses are conflicting (Fig. 5, Table 3). Phylogenetic tree inconsistencies are common among mammals, especially due to important evolutionary events (Rokas and Chatzimanolis 2008). These discrepancies can be attributed to factors such as inadequate gene sampling, hybridization events, gene introgression, or horizontal transfer. Although these findings provide enough evidence to consider B. caspica as an independent species (Fig. 5, Table 3), our understanding of its evolutionary relationship with other Barbastella species remains limited. Furthermore, the Central Asian species is named B. walteri (Smirnov et al., 2021). Hence, to obtain more comprehensive information, it is necessary to explore the genomic aspects of all species rather than confining our study solely to partial genes.

Previous reports have indicated that B. caspica is distributed from the Caucasus region through Iran to Tajikistan (Kruskop 2015), excluding China. This report has expanded our understanding of the geographic distribution of B. caspica. Combining these findings with previous research, we infer that the eastern edge of the B. caspica distribution extends to Xinjiang, China. Before this discovery, only two species of barbastelles (B. darjelingensis and B. beijingensis) had been documented in China, and B. darjelingensis was found exclusively in Xinjiang. Therefore, our report adds an additional species of barbastelle bats to the Chinese biodiversity.

This study highlights the presence of B. caspica in Xinjiang, China, for the first time and presents the first complete assembly of the mitochondrial genome, providing valuable genetic resources for investigating inter- and intraspecific evolutionary relationships. In addition, we describe for the first time free-flight echolocation calls, possibly of type-I sounds omitted through the mouth. Taking the collection site of our specimen of B. caspica into account, it is necessary to conduct further ecological and genetic studies at the population level on a whole distributional scale.

We express our sincere appreciation to subject editor and reviewer for their constructive comments and suggestions on our manuscript. We thank Chai Guanghou and Zhang Yan for their indispensable support during the two-month field sampling period. We acknowledge financial support from the National Natural Science Foundation of China (grant no. 32260328). We thank the editor, Wieslaw Bogdanowicz, and the reviewers for their constructive comments on the manuscript.