Identification of a novel adenovirus in liver tissue sample of the Great Himalayan leaf-nosed bat (Hipposideros armiger)

Abstract

Bats are important reservoirs for many zoonotic viruses. To explore and monitor potential novel viruses carried by bats, 21 liver samples of bats (Hipposideros armiger) were collected from Yunnan Province in southern China. Only one (4.8%) of all models was detected with adenovirus. The whole genome strain obtained by the viral metagenomics method combined with PCR was temporarily named YN01. The complete genome of YN01 was 37,676 bp, with a G + C content of 55.20% and 28 open reading frames. Phylogenetic analysis indicated that the strain YN01 can be classified as genus Mastadenovirus and was the most similar to the adenovirus isolated from Rhinolophus sinicus in China in 2016. The analysis is needed to verify the possibility of cross-species transmission. This virological investigation has increased our understanding of the ecology of bat-borne viruses in this area and provided a reference for possible future infectious diseases.

Introduction

With the utilization and evolvement of sequencing technology, pathogen detection methods, and changes in the climate environment caused by human activities, novel pathogens are constantly being discovered and identified. The outbreak of SARS-CoV-2 pneumonia in late 2019 and its associated animal hosts have raised concerns about zoonotic infections [1, 2]. Bats are important hosts for many zoonotic viruses and may carry a variety of significantly differentiated and unacknowledged novel viruses [3], including adenovirus, Ebola virus, rabies virus, coronavirus, and astrovirus [4–8]. However, there are more than 2005 known bat species worldwide, exhibiting significant social populations, mobility, the ability to colonize in human-made environments, and a diverse range of species [9]. Therefore, dynamic detection of viruses housed in bats can help us better understand the ecology of bat-borne viruses in the area and provide a valuable baseline for preventing and treating possible viral diseases in the future.

Adenoviruses are non-segmented, linear double-stranded DNA viruses of 35–36 kb [10]. Its natural hosts are mainly vertebrates, including humans [11], monkeys [12], bats [13], and bovines [14]. In humans, adenoviruses can cause respiratory diseases, bronchitis, pneumonia, cystitis, and gastroenteritis [15]. In animals, adenoviruses may cause hepatitis and respiratory and intestinal diseases [16]. According to the classification principles of the International Committee for the Taxonomy of Viruses (ICTV), members of the family Adenoviridae are mainly divided into six genera, namely Atadenovirus, Aviadenovirus, Ichtadenovirus, Mastadenovirus, Siadenovirus, and Testadenovirus. Among them, Mastadenovirus contains the most significant number of species and only infects mammals [17]. Adenoviruses have been found in more than 45 bat species worldwide, and there have been reports of cross-species transmission [17–20]. Microbes carried by bats can mutate and recombine to create new viruses and cause epidemics in animals and even humans [21]. In addition, abundant novel viruses discovered in different bat hosts have also been reported, which indicates that we know little about the viral diversity of bat hosts [22–24]. Fortunately, the development of viral metagenomics allows us to explore these unknown viruses.

Here, a metagenomic approach was used to explore liver tissue samples of the Great Himalayan leaf-nosed bat in Yunnan Province in southern China. This research will help us investigate potential vertebrate viruses from bats and anatomize their genetic relationships.

Materials and methods

Sample collection and storage

In January 2017, 21 bats were captured in Mengla County, Xishuangbanna Prefecture, South China’s Yunnan Province, using the net method. Fresh liver tissue specimens were collected from these bats using sterile surgical instruments after they were safely immobilized and anesthetized. The extracted liver tissue samples were promptly placed into test tubes containing RNA stabilizers, labeled with relevant information, transported to the laboratory through a cold chain, and stored at − 80 ℃ for subsequent analysis.

Sample pretreatment and nucleic acid extraction

Liver tissue samples were removed from the − 80 ℃ refrigerator and refreshed to room temperature. One milliliter of Dulbecco’s phosphate-cushioned saline (DPBS) was added to each sample and swayed vigorously with an oscillator for 5 min [25]. After ultra-fast centrifugation at low temperatures (4 ℃, 12,000 g, 5 min), the liver tissue fragments were centrifuged to the bottom of the tube, and the upper part of the liquid was absorbed into a new 1.5-mL centrifuge tube. The retained supernatant was filtered through a 0.45-μm pore filter (purchased from Merck Millipore) in a biosafety cabinet. After centrifugation for 5 min (4 ℃, 12,000 g), the unprotected nucleic acid was digested by digestive enzymes at 37 ℃ for 60 min. Mix the mixture upside down once during the digestion reaction for 30 min [26]. Samples after pre-treatment and enzyme-digested were extracted per the instructions of the QIAamp MinElute Virus Spin Kit (purchased from Qiagen). The RNA from the total nucleic acid was then synthesized into single-stranded DNA using a reverse transcription PCR reaction employing Life Technology’s Super Script III. The double-stranded DNA was then synthesized by PCR using the Klenow enzyme (purchased from New England Biolabs).

Library preparation and sequencing

Three libraries (each containing seven samples) were constructed from 21 liver tissue samples according to Illumina’s Nextera XT DNA Sample Preparation Kit. Primer dimers and inorganic salts were removed using the QIAquick PCR Purification Kit (purchased from Qiagen). The purified library nucleic acid was screened into fragments in the range of 250–500 bp for sequencing on the Miseq Illumina platform with 250 bp pair-ended sequencing, and each library paired ends with double barcodes [27].

Raw data quality control, assembly, and distribution

After the sequencing of the Illumina Miseq platform was completed, the 250 bp paired-end reads were debarcoded using Illumina’s professional software (Vendor) to obtain the raw data of the library sequencing for further bioinformatics analysis: Process sequencing data using an internal analytics pipeline running on a 32-node Linux cluster. Bowtie2 software was used to eliminate human and other prokaryotic and eukaryotic gene sequences in the library. The low-quality tail produced by sequencing was removed with a Phred score of 20 as the lower limit. The joints and primers from the end of the sequence were trimmed using the VecScreen’s internal parameters. Remove cloned sequence: If nucleotides at sequence positions 5 to 55 were the same, the sequences were considered to have been read repeatedly, leaving only one repeat. The cleaned sequence was assembled using soapdenovo2 software with the default Kmer parameter [28]. The Contig and Singlet sequences obtained by concatenation matched the known virus proteome database through BLASTx. The E-value cutoff of < 10⁻⁵. The crucial hit virus sequences matched by BLASTx are then compared with the internal non-viral non-redundant (NVNR) proteome database to rule out false-positive viral hits.

Genome analysis

The DIAMOND software generated virus sequence results in DAA format files, which were subsequently converted into rma6 form files using MEGAN software. The entire genome was then analyzed using Geneious v11.1.2 software. The “De Novo Assemble” function with Low Sensitivity/Fastest parameters combines Contigs that may come from the same genome without overlapping groups to obtain a certain extended maximum Contig. Then, by “Map to Reference,” the original data is mapped to the library, and the maximum Contig is obtained by extension. Those prolonged contigs with major non-structural and structural proteins and with a length > 5000 bp were subjected to thorough characterizing. Putative viral open reading frames (ORFs) were predicted using the built-in parameters of Geneious software [29]. The annotation of ORF was signed by BLASTx comparison between NCBI and the internal non-redundant protein database. The E-value cutoff of < 10⁻⁵.

Phylogenetic analysis

Based on the amino acid sequences of the characteristic conserved region of the novel adenovirus obtained in this study, corresponding amino acid sequences of similar strains and representative strains were downloaded from the NCBI GenBank database to construct phylogenetic analysis. The downloaded amino acid sequence and the amino acid sequence of the new adenovirus were integrated into a fasta format file. Then, the MUSCLE [30] default setting parameter in MEGA v7.0 was used for alignment between sequences. The E-INS-I algorithm in MAFFT v7.3.1 [31] was used to optimize the results further. Moreover, Geneious was used for nucleotide pruning and matching. Finally, the modified alignment results were imported into MrBayes 3.2.7 to construct the phylogenetic tree [32]. The setup procedure follows: The Markov chain runs for a maximum of 1 million generations, collecting samples every 50 generations and discarding the first 25% of the Markov chain Monte Carlo (mcmc) samples. In order to verify the correctness of all Bayesian inference trees, the MEGA v7.0 Maximum-Likelihood method was used to construct phylogenetic trees [30].

PCR screening

Twenty-one liver samples were screened for novel adenovirus by PCR reaction to facilitate the investigation of the prevalence. In order to obtain the complete genome sequence of the novel adenovirus, different primer sequences were designed, with specific descriptions shown in Table 1.

Table 1.

Primers used for specific PCR confirmation

Primer	Targeted protein	Sequence (5′–3′)	Fragment size (bp)	Annealing temperature (°C)
Protein V sense	Protein V	GCGCGTCCGAATTCTTCTTC	975	60
Protein V antisense		CTACCTGCGGCATGATGCTA
pVI sense	pVI	GCATTCACGGAGCGATTGAC	820	60
pVI antisense		GAGAGCGTTCTGTGGTCACA
E3-1 sense	E3-1	CAAAACTGGCCTTGGTGGTG	729	60
E3-1 antisense		GAGTGAAAGCAGGCAAAGGC
E3-2 sense	E3-2	GCACCCCCAGAGCATAGATC	435	60
E3-2 antisense		GCACCCCCAGAGCATAGATC
Penton protein sense	Penton protein	CAGCACCAAGGACAACAACG	1309	60
Penton protein antisense		CGATTGCCGGCATCTTGAAG
Protein IIIa sense	Protein IIIa	GCGGAGAGCACGTATAACGA	661	60
Protein IIIa antisense		CGTTGTTGTCCTTGGTGCTG

Data availability

The sequence obtained in this study has been deposited in the GenBank database under accession number OK032372.

Results

Discovery of a novel adenovirus in bats

The BLASTx results containing sequencing data from all three libraries generated by DIAMOND (DAA format) were imported into Megan6 software to generate rma6 format files, which were used for classification and visualization. The results showed that adenovirus-related reads were detected in only one library. The adenovirus’ complete genome was obtained using the assembly procedure in Geneious 11.1.2 combined with PCR (Table 1). This adenovirus strain was temporarily named YN01. The total genome size of the strain YN01 was 37,676 base pairs (bp), with a G + C content of 55.20%, and owns 28 open reading frames (ORFs). The genome organization of YN01 resembles that of the Mastadenovirus member in that it contains a 104 bp inverted terminal repeat (ITRs) at each end (Fig. 1). BLASTn searching in the GenBank database with the full-length sequence indicated that the best match of the strain YN01 was Bat mastadenovirus WIV9 (KT698853), sharing 84.82% nucleotide sequence identity (Table 2). In order to investigate the prevalence of YN01, nested PCR was performed to screen positive samples. The PCR thermal cycle procedure was set as follows: 95 ℃ 5 min, 95 ℃ 30 s 30 cycles, 55 ℃ (first round) or 60 ℃ (second round) the 30 s, 72 ℃ 40 s or 30 s, and finally 72 ℃ for 5 min. The premixed enzyme rTaq (TaKaRa) was used in the reaction system. Primers used for PCR were designed by using Geneious 11.1.2. A 443-bp and 288-bp region of the DNA polymerase gene of YN01 was amplified with the two sets of primers. The outer primer set Bade-Ws: 5′-TGTATGAAGAGGATCGCGGC-3′ and Bade-Wx: 5′-GAGGGTGCGTTTGAGACTCA-3′ and the inner primer set Bade-Ns: 5′- TGGAAACCCGAGGTAAGCAC-3′ and Bade-Nx: 5′-GGGCATCGGCATAGTAGCAA-3′, the results showed that only one sample was positive, with a total positive rate of 4.8% (1/21) in Hipposideros armiger.

Fig. 1.

Genomic organization of the strain YN01 identified in the bats

Table 2.

Similarity analysis between the strain YN01 and other adenoviruses

Adenovirus reference strains	GenBank accession No	Genome length	G + C content (%)	Identity (%)
Simian adenovirus 18	FJ025931	31,967	61.4	77.04
Titi monkey adenovirus ECC-2011	HQ913600	36,838	59.6	76.42
Bat mastadenovirus WIV9	KT698853	37,545	55.5	84.82
Bat mastadenovirus WIV10	KT698854	37,556	55.2	84.79
Bat mastadenovirus WIV11	KT698855	38,073	55	184.79
Squirrel monkey adenovirus isolate AdV-1	MN017133	37,431	61.6	284.79

The total genome size of the strain YN01 was 37,676 base pairs (bp), with a G + C content of 55.20%, and owns 28 open reading frames (ORFs). There is a 104-bp inverted terminal repeat (ITRs) at both ends of the YN01 genome.

Phylogenetic and recombination analysis of YN01

Phylogenetic analysis based on the conserved DNA polymerase amino acid sequence of adenovirus indicated that the strain YN01 as a separate clade was close to other currently known bat-related mastadenoviruses (WIV9-11). All these three viruses were isolated from Rhinolophus sinicus in China (Fig. 2). Therefore, the strain YN01 can be classified as the genus Mastadenovirus.

Fig. 2.

Phylogenetic analysis of the strain YN01 identified in the bats

Bayesian inference tree based on amino acid sequences of DNA polymerase of viruses belonging to the family Adenoviridae identified here. The virus found in this study is marked with a red line within the tree. Each scale bar indicates 0.3 amino acid substitutions per site.

To further explore the evolutionary relationship between the strain YN01 and other viruses, we performed recombination analysis (Fig. 3). Recombination analysis of the strain YN01 and Simian adenovirus 18 (FJ025931), Titi monkey adenovirus ECC-2011 (HQ913600), Bat mastadenovirus WIV9-11 (KT698853-55), and Squirrel monkey adenovirus isolate AdV-1 (MN017133) was performed using SimPlot. However, the results did not reveal the occurrence of the prominent reorganization event.

Fig. 3.

Recombination analysis between the strain YN01 and other adenoviruses

Each colored line indicates a group of adenovirus strains, with OK032372-Bat-Adenovirus-strain-YN01 as the query sequence.

Discussion

Bats have become one of the critical virus repositories because, like other animals, they often carry multiple pathogens such as viruses, mycoplasma, chlamydia, bacteria, and a variety of parasites. More emerging RNA viruses related to bats have been considered in the past few years. In contrast, there are few reports on identifying DNA viruses from bats [8]. As double-stranded DNA viruses, adenoviruses can infect almost all major vertebrate species. To date, dozens of different species of bat adenovirus (BtAdV A to J) have been delineated worldwide. Since adenovirus was first isolated from the cell line of the Ryukyu flying fox (Pteropus et al.) in 2006 [33], subsequently, more adenovirus strains have been isolated from bats, including the first full-length sequenced BtAdV-3 strain TJM from a Rickett’s big-footed bat (Myotis ricketti) [34]. In China, studies have identified adenoviruses from the Rickett’s big-footed bats (Myotis ricketti), the insectivorous common bent-wing (Miniopterus schreibersii), and the fruit-eating Leschnault’s rousette bat (Rousettus leschenaulti) [35–37]. In addition, several other studies from Brazil [38], North America [39], South Africa [17], Hungary [40], Kenya [41], and Germany [42] have shown large genetic virus diversity in bats.

As an essential component of the ecosystem, bats contribute significantly to biodiversity and play a crucial role in seed and pollen transmission, acting as essential pollinators. However, in certain circumstances, bats may carry pathogens that can cause disease in other animals and, in some cases, even transmit to humans [43]. Influenza A virus [44] and Nipah virus [4] are examples. Bats can also pose a risk to water supplies; parasitic organisms on bats, such as lice and ticks, may serve as vital vectors for pathogen transmission, potentially infecting poultry kept by humans and contaminating the human living environment. Additionally, some small mammals or birds may use bat burrows or habitats as their homes, forming a symbiotic relationship with bats and contributing to the spread of pathogens. Furthermore, some plants rely on bats for pollination and seed dispersal, presenting potential sources of infection. Nevertheless, existing findings do not provide conclusive evidence regarding whether the adenovirus carried by bats can directly cause diseases in humans or other animals. The underlying pathogenic mechanism requires further in-depth study.

This study used metagenomic methods to analyze 21 liver samples of bats (Hipposideros armiger) collected in Yunnan Province, Mainland China. A novel adenovirus with an entire genomic organization was identified, which expanded our understanding of DNA viruses in limited bat populations to some extent. The full-length genome showed that the encoding structure of YN01 was similar to that of most identified adenoviruses. The G + C content of the YN01 genome was 55.20%, consistent with other recognized mastadenoviruses (43.6 to 63.9%) [17]. At the same time, YN01 also contains 104-bp inverted terminal repeats, and this particular structure exists in all known adenoviruses. The above results indicated that both ends of the YN01 genome were successfully sequenced in this study. Phylogenetic analysis showed that the strain YN01, as an independent clade, did not cluster on the same clade as its most similar WIV9-11, so YN01 may be a potential novel species of genus Mastadenovirus. Considering the broad host range of bat adenoviruses and the increasing contact between bats and humans and other poultry, further dynamic monitoring of these viruses is necessary for the future.

Funding

This work was supported by the National Key Research and Development Programs of China under No. 2017YFC1200201, Jiangsu Provincial Key Research and Development Projects No. BE2017693, National Natural Science Foundation of China No. 81741062, and Innovation and entrepreneurship training program for college students in Jiangsu Province.

Declarations

Animal and human rights statement

The study was approved by the Jiangsu University Ethics Committee on the use of animals and complied with Chinese ethics laws and regulations.

Conflict of interest

The authors declare no competing interests.