Background

Nearly 60% of emerging infectious diseases in humans are zoonotic, with up to 70% of them being found to originate from wildlife [1]. Bats have been identified as natural reservoirs of many viruses. Some of these viruses cause outbreaks of severe disease in humans [2], including the Ebola virus, the lyssavirus, the severe acute respiratory syndrome coronavirus, and henipaviruses [3]. Interestingly, these viruses rarely cause apparent clinical signs in bats [2]. Bats possess unique characteristics that may contribute to their ability to act as a major natural reservoir for viruses, including a high level of species diversity, a long lifespan, a high population density, and high levels of spatial mobility [4].

Previous studies mainly focused on bat-borne viruses that are transmitted via respiratory droplets [3]. However, in recent years, several hepatitis virus-related sequences, including those associated with hepadnaviruses, hepeviruses, hepatoviruses, and hepaciviruses, have been found in bats across the globe, indicating the importance of bats as the natural reservoirs of these viruses [5–9].

Hepatitis viruses include hepatitis viruses A, B, C, D, and E, which cause human hepatitis diseases. Hepatitis A virus (HAV) is classified as belonging to the genus Hepatovirus in the family Picornaviridae. Hepatitis B virus (HBV) is classified as belonging to the genus Orthohepadnavirus in the family Hepadnaviridae. Hepatitis C virus (HCV) is classified as belonging to the genus Hepacivirus in the family Flaviriridae. Hepatitis D virus (HDV) is considered to be a subviral satellite because it can only propagate in the presence of HBV. Hepatitis E virus (HEV) is classified as belonging to the genus Orthohepevirus in the family Hepeviridae. Hepatovirus-related sequences have been identified in 13 species of bat collected in North America, Europe, and Africa [5]. Hepadnavirus-related sequences have been discovered in five species of bat collected in Panama, Gabon, Myanmar, and China [6, 8–10]. Highly diverse hepacivirus-related sequences have been detected in 20 species of bat across the world [11]. Hepevirus-related sequences have been discovered in bats in Ghana, Panama, and Germany [7]. These results indicate that bats may be important reservoirs of these hepatitis viruses (Table 1).

Table 1 Hepatitis virus-like sequences detected in bats
Full size table

There are around 120 species of bat in China; however, only limited information has been reported regarding the hepatitis viruses, a novel Orthohepadnavirus in pomona roundleaf bats from Yunnan province was identified in 2015 [9]. In this study, we report the discovery of four novel hepadnaviruses and a hepevirus in our archived bat liver samples that had been collected from several bat species and various geographical regions in China.

Methods

Samples

A total of 78 liver tissue samples were collected from dead bats caused by accident during sampling, which comprised two families, five genera, and 17 species, and used for virus screening (Table 2). Different tissues (heart, liver, spleen, lung, kidney, brain and intestine) were collected separately and used for analyzing virus tissue tropism. The animals were firstly identified based on their morphology and then the species that they belonged to were further confirmed using DNA sequencing of the mitochondrial cytochrome b (CytB) gene following previously described methods [12].

Table 2 Detection of hepadnavirus and hepevirus in bats in China between 2008 and 2013
Full size table

RNA extraction and PCR

RNA was extracted from tissue using the QIAamp Viral RNA Mini Kit (Qiagen, Hilden, Germany) following manufacturer’s instructions, and cDNA was synthesized using Moloney Murine Leukemia Virus (M-MLV) Reverse Transcriptase (Promega, Madison, WI, USA). The extracted RNA from liver was tested by nested or heminested reverse transcription PCR (RT-PCR) using degenerate primers based on the conserved domain of the RNA-dependent RNA polymerase (RdRp) gene of viruses in the genus Hepatovirus, the polymerase gene of viruses in the family Hepadnaviridae, the RdRp gene of viruses in the genus Hepacivirus [11], and the RdRp gene of viruses in the family Hepeviridae [7] (Table 3). Standard precautions were taken to avoid contamination of the PCR procedure, and no false-positives were observed in the negative controls. The PCR products underwent gel purification with MinElute Gel Extraction Kit (Qiagen, Germany) and they were sequenced with both forward and reverse primers using the 3100 Sequencer (ABI, Waltham, MA, USA).

Table 3 Primers used for virus RT-PCR screening and virus quantification
Full size table

Genomic sequencing

The complete genomic sequences of one hepadnavirus strain and one hepevirus strain were amplified using PCR with degenerate primers (the primers are available upon request). The genome ends were amplified using a 5′-Full RACE Kit (TaKaRa, Japan). The PCR products underwent gel purification with MinElute Gel Extraction Kit (Qiagen, Germany) and they were sequenced with both forward and reverse primers using the 3100 Sequencer. The sequencing chromatograms were inspected for overlapping multicolor peaks, which are an indicator of sequence heterogeneity in the amplicons. The PCR products were cloned using the pGEM-T Easy Vector System (Promega, Germany) and at least three clones for each PCR fragment were sequenced to obtain a consensus sequence.

Sequence analysis

The preliminary sequence management and analysis were carried out using Geneious version 9.1.3 (Biomatters Ltd., Auckland, New Zealand) and the sequence alignment and editing were performed using MAFFT [13]. The phylogenetic analysis of hepadnavirus used the neighbor-joining (NJ) method with Hasegawa-Kishino-Yano substitution model and complete deletion option and hepevirus used the maximum-likelihood (ML) method with the nucleotide percentage distance substitution matrix and the complete deletion option in MEGA version 7 [14]. The sequences and GenBank accession numbers of the representative viruses in the families Hepadnaviridae and Hepeviridae used in the phylogenetic analyses are presented in Figs. 1 and 2.

Fig. 1
figure 1

Phylogenetic analysis of bat hepadnavirus based on the full-length genomic sequences. Maximum likelihood phylogenetic tree was constructed based on the complete genomes of BtHBVRs3364 (in bold) and representative members of the family Hepadnaviridae using the Hasegawa-Kishino-Yano substitution model and complete deletion option in MEGA version 7. The values at the nodes indicate the bootstrap values (using 1,000 replications). The branches are labeled with the strain designation, the host species, and the GenBank accession number. The classification of the family Hepadnaviridae is indicated on the right

Full size image
Fig. 2
figure 2

Phylogenetic analysis of bat hepevirus based on the full-length genomic sequences. Neighbor joining phylogenetic tree was constructed based on the alignment of the complete genomes of BtHEVMd2350 (in bold) and representative members of the family Hepeviridae using nucleotide percentage distance substitution matrix and complete deletion option implemented in MEGA version 7. The values at nodes indicate the bootstrap values (using 1,000 replications). The branches are labeled with the strain designation, the host species, and the GenBank accession number. The classification of the family Hepeviridae is indicated on the right

Full size image

Quantification real-time PCR

Virus load of bat hepevirus and hepadnaviruese of different tissues was measured by using photometrically quantified in vitro RNA transcripts and specific real-time RT-PCR primers (Table 3). Quantification was done by using 5 μL of RNA extract, 300 nM each primer, using the One Step SYBR PrimeScript™ PLUS RT-PCR Kit (TaKaRa, Japan). Cycling in a Biorad CFX Connect instrument involved the following steps: 42 °C for 5 min, 95 °C for 10 s, and 40 cycles of 95 °C 5 s and 60 °C 20 s with measurement of fluorescence.

Results

Detection of four hepadnaviruses and a hepevirus in bat liver samples

Among the 78 bat liver samples, four were positive for hepadnavirus from Jinning city, Yunnan province and only one was positive for hepevirus from Xianning city, Hubei province (Fig. 3). However, none were positive for hepatovirus or hepacivirus. The nucleotide sequences of the four novel hepadnaviruses and the hepevirus described in this study are available from GenBank under the accession numbers KX513949–KX513953.

Fig. 3
figure 3

Representation map of China and Yunnan province. Circle indicates the sampling site where the hepevirus (BtHEVMd2350) was detected. Triangles indicates the sampling sites where the hepadnaviruses were detected

Full size image

Sequence analysis of the bat hepadnavirus

All four of the hepadnavirus-positive samples were from horseshoe bats, two each from R. sinicus (designated BtHBVRs3364 and BtHBVRs3366) and R. affinis (designated BtHBVRa4325 and BtHBVRa4328) (Table 2). The four partial polymerase gene sequences had 92.1–97.5% nucleotide sequence identity and they were found to be closely related to the roundleaf bat hepadnavirus from Yunnan province, China, with nucleotide identities of 88.8–95.5% [9].

The full-length genomic sequence of a sample from R. sinicus (designated bat HBV Rs3364, or BtHBVRs3364) was determined and it was found to have a length of 3,272 nucleotides. The virus has an identical genomic organization to other hepadnaviruses, with four open reading frames (ORFs) encoding the surface (S), polymerase (P), core (C), and X proteins. In addition, the typical direct repeat (DR) sequences for viral genome replication and the secondary structure Ɛ–loops for viral reverse transcription were present in the BtHBVRs3364 genome. A detailed comparison of the full-length genomic sequence and the ORFs of the virus with other known hepadnaviruses is shown in Table 4. The results showed that the four genes of BtHBVRs3364 have the highest degree of identity with the roundleaf bat hepadnavirus from Yunnan province, at both the nucleotide and the amino acid levels [9]. Notably, we found large differences between BtHBVRs3364 and other hepadnaviruses from the African horseshoe and roundleaf bats, the long-fingered bat from Myanmar, and the tent-making bat from Panama.

Table 4 Nucleotide and amino acid sequence identity between BtHBVRs3364 and representative orthohepadnavirus strainsa
Full size table

Phylogenetic analysis of the bat hepadnavirus

A phylogenetic tree was constructed based on the alignment of the full-length genomic sequence of BtHBVRs3364 with those of representative hepadnavirus strains available in GenBank. As shown in Fig. 1, the previously reported bat hepadnaviruses formed three clusters, with clear specificities for particular hosts. Although BtHBVRs3364 clustered with the bat hepadnaviruses, it formed an independent branch. Interestingly, the BtHBVRs3364 detected in the horseshoe bat is phylogenetically closer to viruses from the Asian roundleaf bat compared to viruses from the African horseshoe bat, despite the fact that it was found in an Asian horseshoe bat.

Sequence analysis of the bat hepevirus

One sample found in the whiskered bat, M. davidii, from Hubei province was positive for hepevirus (designated bat HEV Md2350, or BtHEVMd2350). The genomic sequence of BtHEVMd2350 was found to have a length of 6,607 nucleotides (excluding the poly(A) tail at the 3′ end). This is slightly shorter than BS7 (which has a genomic sequence length of 6,671 nucleotides), the only reported bat hepevirus with a fully sequenced genome, which was identified from the serotine bat, Eptesicus serotinus, in Germany [7]. BtHEVMd2350 was found to have a 5′ untranslated region (UTR) of 33 nucleotides and a 3′ UTR of 76 nucleotides. The three unique ORFs that are found in other members of the family Hepeviridae were also found in BtHEVMd2350: ORF1 encodes a nonstructural polyprotein that includes the RdRp, ORF2 encodes the capsid protein, and ORF3 encodes a multifunctional protein. Notably, most of the elements and domains characterized in BS7 could be found in BtHEVMd2350, but with a high level of divergence (Table 5).

Table 5 Nucleotide and amino acid sequence identity between BtHEVMd2350 and representative hepevirus strainsa
Full size table

Phylogenetic analysis of the bat hepevirus

A phylogenetic tree was constructed based on the alignment of the full-length genomic sequence of BtHEVMd2350 with those of representative full-length hepevirus genomic sequences (Fig. 2). The results showed that bat hepeviruses (BtHEVMd2350 and BS7) cluster into a separate monophyletic clade within the family Hepeviridae.

Quantification of novel viruses

Viral load detected by qPCR in different tissues were presented in the Fig. 4. The highest viral load of the BtHEVMd2350 was found in the liver (1.9 × 1010 RNA copies per gram of tissue) and followed by spleen (7.3 × 108 RNA copies per gram of tissue), intestine and kidney, but not detectable in the brain. For bat hepadnavirus, the highest viral load was found in the liver of BtHBVRs3364 (2.0 × 1010 RNA copies per gram of tissue), the virus load of tissues of BtHBVRs3366, BtHBVRa4325, and BtHBVRa4328 were relatively similar (medien, 6.2 × 106 RNA copies per gram of tissue; range, 4.9 × 105 to 2.7 × 1010 RNA copies per gram of tissue).

Fig. 4
figure 4

Tissue distribution of bat hepevirus and hepadnaviruses. Virus concentrations assessed by virus-specific one-step real-time RT-PCR using quantified in vitro-transcribed RNA controls

Full size image

Conclusions and discussion

Since the discovery of genetically diverse hepatitis virus-related sequences in bats, bats have been considered to be important natural reservoirs for hepatitis viruses, and potential sources of human diseases [10]. However, these hypotheses need to be proved by screening more bat samples from across the globe for hepatitis viruses. In this study, we screened for hepatitis viruses in bats from China and discovered four novel hepadnaviruses circulating in two species of horseshoe bat in Jinning city, Yunnan province and one hepevirus in the whiskered bat M. davidii in Xianning city, Hubei province. The full-length genomic sequences of one of the two hepadnaviruses from R. sinicus and the hepevirus from M. davidii were determined.

The phylogenetic analysis indicates that the bat hepadnavirus found in this study is closely related to roundleaf bat hepadnaviruses, which were discovered in Pu’er city, Yunnan province in 2011 [9], but shows remarkable divergence when compared to the African horseshoe bat, despite the fact that it was found in an Asian horseshoe bat. A similar phylogenetic relationship was found between hepadnaviruses from the African roundleaf bat and the African horseshoe bat [6], indicating the separate evolution of these viruses and their hosts.

Regarding the bat hepevirus, the phylogenetic analysis indicates that the known bat hepeviruses are highly divergent from other mammalian hepeviruses and that they form an independent branch in the family Hepeviridae. According to the latest proposal of the ICTV in 2016, amino acid distances of concatenated ORF1 and ORF2 (lacking hypervariable regions) greater than 0.088 could then act as threshold to demarcate intra- and inter- genotype distances [15]. The hepevirus detected in the whiskered bat, M. davidii, and that found in the German serotine bat, E. serotinus (the only reported bat hepevirus with a full-length genome) shared significant diversity from both nucleotide and amino acid levels, we propose that they can be grouped into the species Orthohepevirus D which is divided into two genotypes: D1 and D2.

Our results provide further evidence to support the theory regarding the long-term co-evolution of hepadnaviruses and hepeviruses with their hosts, and the theory that bats act as major natural reservoirs for these hepatitis viruses. Our results have limitations due to the small sample size used, which was a result of the protection of bat populations in China, as bats play important roles in the pollination of plants and in pest control, as they feed on insects. However, based on our discovery of hepatitis viruses in bats, it is expected that there are many more hepatitis viruses circulating in numerous bat species and in various geographic regions. In order to obtain larger sample sizes, non-invasive methods of virus detection should be considered for future studies.