Abstract
Introduction:
Viruses in water bodies pose potential threats to human health and ecological environments. Therefore, exploring viral diversity in human-associated lakes and domestic wastewater is crucial.
Methods:
In this study, we performed comparative viromic analysis of lake and domestic wastewater samples collected from the same water system in Wenzhou, China.
Results:
A total of 13,708,121 sequencing reads were classified into 30 viral families, including viruses associated with animals, plants, and algae. Alpha diversity analysis showed that the lake exhibited higher viral richness and evenness than domestic wastewater. Notably, two ginkgo-associated viruses potentially infecting arthropods were detected in the lake: Ginkgo biloba picorna-like virus and Ginkgo biloba dicistrovirus. These clustered evolutionarily with arthropod-infecting viruses, suggesting a potential capacity to infect insects. Furthermore, Swine acute diarrhea syndrome coronavirus (SADS-CoV) sequences were detected in urban water environments and clustered with previously reported SADS-CoV strains. Evolutionarily, this SADS-CoV clade is closely related to the bat coronavirus HKU2, sharing the same Rhinacovirus.
Discussion:
These findings expand our understanding of the aquatic ecosystem virome.
1 Introduction
As a vital biological entity, viruses have a profound impact on regulating the structure of microbial communities, driving biogeochemical cycles, and regulating host evolution (; ). Recently, the emergence of viral metagenomics has provided a powerful tool to explore the vast viral diversity in aquatic ecosystems. However, aquatic environments are not isolated. Through rainwater runoff (), agricultural drainage (), and wastewater discharge (), rivers and lakes continuously receive exogenous viruses from terrestrial, human, and animal sources. Aquatic ecosystems are reservoirs of viral genetic diversity.
Aquatic environments frequently receive viral inputs from surrounding human activities and terrestrial ecosystems. Ginkgo biloba picorna-like virus and Ginkgo biloba dicistrovirus, members of the order Picornavirales, have previously been reported only in terrestrial environments (), while their distribution and ecological significance in aquatic environments remain unknown. Concurrently, most human pathogens (e.g., adenoviruses and coronaviruses) have been widely used as wastewater indicator viruses due to their association with fecal-oral transmission (; ; ). Swine acute diarrhea syndrome coronavirus (SADS-CoV) is a highly pathogenic enteric coronavirus with significant economic impacts on the swine industry (; ). Although this virus has primarily been detected in agricultural settings such as swine farms, the possibility of its introduction into adjacent water bodies through agricultural runoff cannot be overlooked. However, the occurrence of SADS-CoV in natural water bodies, its environmental persistence, and potential public health risks remain inadequately studied to date.
To assess viral diversity and potential risks in the aquatic environment, this study selected two water systems: domestic wastewater, which is a series of complex pollutants generated by daily human activities, and lakes, which are typical freshwater bodies regulated by natural ecological processes. Our main research objectives were to: (1) compare the distribution and diversity of the viromes in the lake and domestic wastewater; (2) investigate the unexpected presence and potential origin of the SADS-CoV in an urban water environment sample. and (3) trace the cross-ecosystem transmission of arthropod-associated Ginkgo viruses. Ultimately, this study aims to broaden our understanding of aquatic RNA viruses from both epidemiological and ecological perspectives.
2 Materials and methods
2.1 Sample collection and pre-treatment
Water samples (1,500 mL) were collected from two distinct sites in September 2024 in Ouhai District, Wenzhou, Zhejiang Province, China.The lake water sample (L1) was collected from an artificial lake with an approximate depth of 2–3 m. The domestic wastewater sample (W2) was collected from a flowing sewer outlet in a mixed residential and commercial area. Both sampling sites are adjacent to areas of frequent human activity (Figure 1). Both samples were transported at 4 °C to the laboratory and processed immediately. Specimens were concentrated as described previously (). Briefly, 1,500 mL of each water sample was centrifuged at 6,000 × g for 20 min at 4 °C to remove large particles and debris. Subsequently, MgCl2 (2.5 M) was added to the supernatant to a final concentration of 0.05 M. The pH was adjusted to 3.5 with 0.5 M hydrochloric acid, imparting a positive charge to the viral particles and facilitating their electrostatic binding to the anionic membrane. These steps were performed on a magnetic stirrer. The solution was filtered through a mixed cellulose ester membrane filter with 0.45 μm (Jukai, Shandong, China) to maximize virus-membrane contact and capture efficiency. The membrane was removed in a biosafety cabinet, cut into small pieces, and placed in a 50 mL sterile centrifuge tube. To elute the adsorbed viruses, 15 mL of 3% beef extract solution was added to the membrane pieces, followed by ultrasonication for 20 min to facilitate the release of viral particles from the membrane. After centrifugation at 3,000 × g for 30 min at 4 °C, the supernatant was filtered through a 0.22 μm filter to remove residual bacteria. The pH of the filtrate was then adjusted back to 7.0 with hydrochloric acid to stabilize the viral particles (; ; ). Finally, the samples were stored at −80 °C and used for RNA extraction.
Figure 1
2.2 RNA extraction, library construction and sequencing
Total RNA was extracted from 1 mL of each concentrated viral sample using the HiPure Viral RNA kit (Magen, Guangzhou, China) following the manufacturer's instructions. The concentration of the extracted RNA was measured using NanoDrop2000 (ThermoFisher, USA) to determine the appropriate input amount for library construction. Extracted RNA was stored at −80 °C until further processing. Total RNA extracts were subjected to library construction using the VAHTS Universal V8 RNA-seq Library Prep Kit for Illumina (Vazyme, Nanjing, China). Ribosomal RNA (rRNA) of eukaryotic cytoplasm, mitochondria, and bacteria origin were removed using the Ribo-Clean rRNA Depletion Kit Mega (Bacteria) (Vazyme, Nanjing, China). The quantity and quality of the final RNA libraries were assessed using a Qubit 4.0 fluorometer (Invitrogen, USA). The RNA library construction was carried out in the laboratory, and the sequencing was performed by Beijing Novogene Technology Co., Ltd. Paired-end (150 bp reads) sequencing for each RNA library was performed on the Illumina HiSeq Xplus platform (Novogene, Beijing, China).
2.3 Bioinformatics analysis
After obtaining the raw data of each sample, the sequencing data (Raw data) were quality-controlled with FASTP (v.0.23.4) () to remove the adapters, low-quality and low-complexity reads, and to obtain high-quality clean reads to ensure the credibility of the results of the subsequent analysis. To remove host-derived reads, clean reads were first mapped against the Kraken 2 standard reference database using Bowtie 2 (v.2.4.2) (). After host read removal, the remaining clean reads were taxonomically classified using Kraken 2 (v.2.0.9), and species-level abundance was estimated using Bracken (). The high-quality reads of each sample were then assembled into longer contigs using MEGAHIT (v.1.1.3) (), which was chosen for its high memory efficiency and suitability for the computational resources available in our laboratory.
2.4 Viral sequence identification and virus quantification
The assembled contigs were first aligned against the NCBI non-redundant protein database (nr) (https://www.ncbi.nlm.nih.gov) using DIAMOND blastx (https://github.com/bbuchfink/diamond) with an E-value threshold of < 1 × 10−5. Contigs annotated as viral were then further validated by aligning against the NCBI nucleotide sequence database (nt) using BLASTn with the same E-value threshold. Only contigs independently annotated as viral by both databases were retained as the final viral sequences, thereby minimizing false positives through cross-validation. Viral quantification was performed by mapping the clean reads back to the identified viral contigs and statistically analyzing the read counts.
2.5 Viral alpha diversity
Alpha diversity of the viral communities was assessed at both the family and genus levels. The Shannon index and Chao1 index were calculated using PAST (v.5.2.1) (). Visualization was generated via ChiPlot (https://www.chiplot.online/).
2.6 RT-PCR verification
Based on the results of contig comparison, RT-PCR was conducted to verify the assembled viral sequences and eliminate false positives. Viral contigs were assembled using SeqMan (v.7.1.0) from DNASTAR, and reference sequences were identified by aligning the assembled contigs against the NCBI nucleotide database using BLASTn. Primers were designed based on these reference sequences using Primer Premier 5 (See the Supplementary Table 1). Total RNA was extracted and purified using the HiPure Viral RNA Kit (Magen, Guangzhou, China) and reverse transcribed into cDNA. PCR amplification was performed under the following conditions: initial denaturation at 94 °C for 5 min; 35 cycles of 94 °C for 30 s, annealing at 50–60 °C for 30 s, and extension at 72 °C, with the extension time determined according to the size of each viral fragment; a final extension at 72 °C for 5 min. Negative controls were included in each run to rule out contamination. PCR products were visualized by agarose gel electrophoresis (Supplementary Figure 1). PCR products were sequenced by Sangon Biotech Co., Ltd. (Shanghai, China). The verified viral sequences were submitted to GenBank (Supplementary Table 2).
2.7 Phylogenetic analyses
Conserved protein-coding regions, including nucleoprotein (N) and other conserved hypothetical protein genes, were used for phylogenetic reconstruction in the respective viral taxa. The nucleotide sequences of the above viruses were subjected to sequence alignment using BLASTn against the NCBI nucleotide database., Reference sequences at the taxonomic family level were downloaded. Multiple sequence alignment was performed using MAFFT (v.7.520) with the E-INS-I algorithm (). After alignment, poorly aligned or low-quality regions were trimmed using TrimAl in TBtools (v.2.420) with the ML_AUTOMATED1 mode (). After trimming, the phylogenetic analysis was conducted using the maximum likelihood method in IQ-TREE (v.2.4.0) () under the GTR+F+R3 model, with 1000 bootstrap replicates. The annotations and modifications were carried out using the online website iTOL () (https://itol.embl.de/itol.cgi).
2.8 Genome annotation
Potential open reading frames (ORFs) were predicted using ORFfinder (https://www.ncbi.nlm.nih.gov/orffinder/with) the standard genetic code and a default minimum ORF length of 75 nt. The longest ORF in each sequence was selected for subsequent analysis. Conserved domains were searched against the InterPro database (https://www.ebi.ac.uk/interpro/search/sequence/), including Pfam and CDD. Signal peptides were predicted using SignalP 6.0 to identify potential secretory regions, and transmembrane helices were predicted using TMHMM 2.0 to identify membrane-spanning regions. Sequence profiles were further searched against the HHpred database (https://toolkit.tuebingen.mpg.de/tools/hhpred) to detect remote homologs. Amino acid conservation patterns among the viral sequences and their reference sequences were analyzed using Skylign (https://skylign.org/) to generate sequence logos and identify conserved residues. Multiple sequence alignments of the deduced amino acid sequences were performed using MEGAX (v.10.2.6). Mutation analysis, including the identification and visualization of amino acid substitutions relative to reference sequences, was conducted using BioAider (v.1.527) (Zhou et al., 2020), and the results were displayed as lollipop plots.
2.9 Host prediction
The Viral Host Predictor (http://host-predict.cvr.gla.ac.uk/) is used to predict the host of RNA viruses, arthropod vectors and arthropod-borne, including Arenaviridae, Astroviridae, Bunyavirales (Feraviridae, Hantaviridae, Jonviridae, Nairoviridae, Peribunyaviridae, Phenuiviridae, and Tospoviridae), Caliciviridae, Coronaviridae, Filoviridae, Flaviviridae, Hepeviridae, Paramyxoviridae, Picornaviridae, Rhabdoviridae, and Togaviridae (). At the same time, the RNAVirHost tool () is used to predict the host of the viral genomes. Visualization was generated via ChiPlot (https://www.chiplot.online/).
3 Results
3.1 Overview of metagenomic sequencing data
After removing low-quality reads and host sequences, a total of clean bases is 4.6 Gb. Both groups were annotated as bacteria, eukaryotes, viruses and unknown sequences. The results indicated that bacteria remained an important component in water bodies. Although other biological categories were also identified, a large number of unknown sequences remain, indicating that many aspects of this field are still unknown and await further exploration (Table 1). In the virus category, the proportion of the lake of which 1891 were viruses, accounting for 0.03%, and that of the domestic wastewater libraries, of which 12560 were viruses, accounting for 0.11% (Figures 2A, B).
Table 1
| Sample | Clean reads | Clean base (Gb) | Bacteria | Eukaryota | Unclassified | Virus | Viral read abundance (%) |
|---|---|---|---|---|---|---|---|
| L1 | 3740332 | 1.49 | 3260663 | 184305 | 2120248 | 1891 | 0.03 |
| W2 | 9967789 | 3.17 | 9101450 | 344030 | 3019948 | 12560 | 0.1 |
Overview of sequencing libraries of different viromes.
The sequencing reads from the two sample groups were mainly annotated as bacteria, eukaryotes, viruses, and unclassified.
Figure 2
Virus discovery and identification were performed by assembling contigs from scratch, and species type and viral differences were statistically analyzed for the two sample groups. The results showed that a species type analysis of the two sample groups was performed using a Venn diagram. The two groups had three types in common: as plants, insects, and mammals. Among them, the lake (L1) had one unique species type, which belonged to the environmental sample, while the domestic wastewater (W2) had two unique types, as mollusc and Homo sapiens (Figure 2C). At the same time, we analyzed the viral species composition of two sets of libraries. The results showed that lake (L1) had the most viral species, with 19 species, while domestic wastewater (W2) had only 12 viral species. Among them, the lake (L1) carried 14 unique viruses, and the domestic wastewater (W2) carried 7 unique viruses. There were 5 shared viruses in the two libraries. This indicated that there were significant differences in the viral community composition of the two libraries (Figure 2C).
TPM is a standardized method for gene expression quantification, enabling accurate comparisons of expression levels within groups, between different groups, and between different viruses. The two groups of samples contained a total of 31 types of viruses. We calculated the relative abundance TPM of all viruses and through the abundance heatmap, which showed significant differences between the two libraries. In L1 library, the abundance of Ginkgo biloba picorna-like virus was the highest, accounting for 36.7%, followed by Ginkgo biloba dicistrovirus and Porcine reproductive and respiratory syndrome virus, accounting for 29.5% and 19.3%. In the W2 library, the Wenzhou picorna-like virus 11 had the highest abundance, accounting for 45.6%, followed by Ginkgo biloba tombusvirus, accounting for 17.9% (Figure 2D).
3.2 Diversity of viral communities
To understand the distribution and abundance of viral families in lake and domestic wastewater, a heat map was constructed based on the viral genome sequences of the two libraries, and classified according to the family level. The viral reads were ultimately annotated to 30 viral families. In DNA viruses, Baculoviridae, Adenoviridae, Rountreeviridae, Straboviridae, Poxviridae, Pachyviridae, Rudiviridae, Kyanoviridae, Orthoherpesviridae, Autographiviridae, Phycodnaviridae and Herelleviridae had double-stranded DNA (dsDNA) genomes; In RNA viruses, Hantaviridae, Paramyxoviridae and Rhabdoviridae had single-stranded negative-sense RNA [ssRNA (-)] genomes; Botourmiaviridae, Togaviridae, Arteriviridae, Virgaviridae, Fiersviridae and Coronaviridae had single-stranded positive-sense RNA [ssRNA(+)] genomes; virus in the Retroviridae had single-stranded RNA reverse transcription (ssRNA-RT); Blumeviridae, Solspiviridae, Steitzviridae and Phenuiviridae were unclassified (Figure 3A).
Figure 3
At the specific families and genus level, Baculoviridae-Betabaculovirus had the highest proportion of all viruses, accounting for 66% in the lake and 95% in domestic wastewater. Rhabdoviridae-Vesiculovirus accounted for 8.99% in the lake and 0.5% in the domestic wastewater. It was clear that there were significant differences in the abundance of bacteriophages and plant viruses in the lake; in contrast, the lowest abundances were found in the domestic wastewater (Figures 3B, C).
Alpha diversity is key to understanding viral diversity and richness in sample libraries. The Chao-1 index can effectively estimate the number of species, while the Shannon index showed the extent of species diversity. The results analyzed the alpha diversity of the two libraries at the family and genus levels. At the family and genus levels, the Shannon index of L1 libraries is higher than W2 libraries. However, the number of W2 species was higher on the Chao-1 index (Figures 3D, E). The results showed that the species richness in the L1 libraries was relatively high, and the uniformity of community distribution was quite significant. In the W2 libraries, the viral diversity showed a relatively high species abundance. The community structure presented dominance by a few species, with a low degree of uniformity and a simple overall structure.
3.3 Phylogenetic analyses of viral sequences
3.3.1 Virgaviridae
Two viral sequences belonging to Virgaviridae, a family of plant-infecting single-stranded RNA viruses (), were assembled from the lake sample (L1). Hubei sediment virgavirus 1 (HBSV1) sequence showed nt identity of 95%−97% to the reference sequence MW897255, and Hubei sediment virgavirus 2 (HBSV2) sequence showed nt identity of 96%−97% to the reference sequence MW897259, both of which were reported from Hubei, China in 2017.Phylogenetic analysis placed both sequences within the unclassified Virgaviridae clade (Figure 4A).
Figure 4
3.3.2 Tombusviridae
A Ginkgo biloba tombusvirus (GBTV) sequence belonging to Tombusviridae, a family of plant-infecting viruses, was assembled from the lake sample. This sequence shared 99.4% nucleotide identity with a previously reported Ginkgo biloba tombusvirus (Genbank: MN729618) from Zhejiang, China in 2019. Phylogenetic analysis placedthis sequence withinthe Tombusvirus and was most closely related to MN729618.1 (Figure 4B).
3.3.3 Narnaviridae
A Ginkgo biloba narna-like virus (GBNV) sequence showing similarity to Narnavirus, currently the only recognized genus within the Narnaviridae according to the latest ICTV classification, was identified in the lake sample and subsequently confirmed by RT-PCR. The result of the nucleotide sequence alignment showed that this sequence shared 98.8% identity with Ginkgo biloba narna-like virus (GenBank: MN862348) and Lactuca sativa narna-like virus (GenBank: MN862349), both previously reported from Jiangsu, China in 2019. Phylogenetic analysis showed that the three sequences are clustered on the same branch (Figure 4C).
3.3.4 Dicistroviridae
A Ginkgo biloba dicistrovirus (GBDV) sequence, belonging to Dicistroviridae, was identified in the lake sample. Through RT-PCR and alignment, this sequence shared 91.88% nt identity with the reference sequence Ginkgo biloba dicistrovirus (GenBank: MN841301) from Jiangsu, China in 2019. Phylogenetic analysis showed that the sequences were clustered in the same branch, within the unclassified Dicistroviridae group (Figure 4D).
3.3.5 Coronaviridae
Two N gene nucleotide sequences of SADS-CoV were assembled from the lake and domestic wastewater samples, respectively. SADS-CoV belongs to the genus Alphacoronavirus within the Coronaviridae. Among the four major structural proteins encoded by coronaviruses, the nucleocapsid (N) protein is highly conserved (; ). Both sequences shared 100% nucleotide identity with MK994936 from Guangdong, China in 2019. Phylogenetic analysis placed both sequences within the Alphacoronavirus, clustering closely with known SADS-CoV strains (Figure 4E).
3.3.6 Picornavirales
In this study, a nucleotide sequence (1388 nt) of Ginkgo biloba picorna-like virus (GBPV) was assembled from the lake library. GBPV belongs to the Picornavirales, a diverse group of non-enveloped positive-sense single-stranded RNA viruses that infect a broad range of eukaryotic hosts (). This sequence shared 92%−94.9% nt identity with Picornavirales sp. (GenBank: PP097442) from Brazil and GBPV-pt112-upi-3 (GenBank: MN832469) from China. The phylogenetic analysis showed that the three sequences clustered on the same branch and belonged to the unclassified Picornaviridae (Figure 4F).
3.4 Virus host prediction
To elucidate the host ranges of the identified virome, host prediction analysis was conducted. Overall, potential hosts were successfully predicted for 62.69% of the viral sequences, while 37.31% remained classified as having unknown hosts (Figure 5A). In lake (L1) and domestic wastewater (W2), the identified viral communities were predominantly predicted to infect Viridiplantae and Invertebrata (Figure 5A). Notably, viruses belonging to the order Picornavirales constituted a major proportion of the community. Further detailed analysis of specific viral species abundance across different host categories revealed distinct distribution patterns (Figure 5B). Viruses targeting Viridiplantae exhibited the highest relative abundance overall. Notably, the highly conserved Ginkgo biloba dicistrovirus and Ginkgo biloba tombusvirus were the most dominant species, showing exceptionally high abundance in both L1 and W2 samples. Most prominently, Swine acute diarrhea syndrome coronavirus displayed unexpectedly high relative abundance in both samples (Figure 5B).
Figure 5
3.5 Analysis of potentially harmful viruses
3.5.1 Characteristics of the Ginkgo virus
Picornavirales members infect diverse hosts globally, driving major agricultural losses (Zhang Y. et al., 2023). In this study, two novel viral sequences from Ginkgo biloba were characterized: Ginkgo biloba picorna-like virus (GBPV) and Ginkgo biloba dicistrovirus (GBDV). GBPV consists of 1312 nucleotides (nt) containing nine ORFs (Supplementary Figure 2). The longest ORF (nt 272–1312) encodes a hypothetical protein sharing 85% amino acid identity and 97% coverage with an unclassified Gingko biloba-like virus (QKK82976) previously reported in China (). Structural predictions revealed that this protein possesses a Sec/SPI-type signal peptide (SP), a transmembrane (TM) helix, and a non-cytoplasmic domain (Figure 6A), strongly suggesting it is a type I transmembrane or secretory protein (Supplementary Figure 3). Specifically, the SP cleavage site is predicted between amino acids 33 and 34 (probability: 0.989).
Figure 6
Further conservation analysis of the GBPV SP and TM domains highlighted a high frequency of hydrophobic residues, such as M and W at positions 1 and 21 in the SP, and Y in the TM domain (Figure 6B), indicating highly conserved hydrophobic functions. The SP sequence exhibited nine mutations (e.g., C16L and C33V), while the TM domain contained ten variations, including N203D/E and A212S (Supplementary Figure 4). These substitutions could potentially alter the protein's structural conformation and biological characteristics.
For GBDV (1044 nt), the single predicted ORF (nt 426-1043) encodes a putative DNA/RNA polymerase (aa 537-857, IPR043502) (Figure 6A). This sequence shares 97.57% amino acid identity and 100% coverage with a Lactuca sativa dicistroviridae (QKI28881). To further elucidate its structure, HHpred analysis was performed. The GBDV protein showed the highest structural homology to the known 6R1I and 8C1N_A proteins, with structural prediction probabilities reaching 99.88% and 99.87%, respectively (Supplementary Figure 5).
3.5.2 Potential host
In this study, by host prediction models based on genomic biases and phylogenetic neighborhood, the results showed that among the potential hosts of Ginkgo biloba dicistrovirus (GBDV, PV874581). Ginkgo biloba picorna-like virus (GBPV, PV925707), insects had a higher probability of matching than other host groups (such as plants, primate), indicating that both viruses are more likely to infect insect hosts (Figures 7A, B). In addition, phylogenetic analysis of GBDV and GBPV showed that the viruses infecting arthropods clustered in the same branch of the phylogenetic tree with high phylogenetic affinity (100%) (Supplementary Figure 6).
Figure 7
3.5.3 Characterization of swine acute diarrhea syndrome coronavirus
We successfully identified swine acute diarrhea syndrome coronavirus (SADS-CoV) sequences in both groups using RT-PCR. A phylogenetic tree was constructed based on the N protein gene predicted from the open reading frame (ORF). Phylogenetic analysis indicated that the two sequences were closely related to SADS-CoV strains circulating in China between 2017 and 2018, showing 98.68% amino acid identify with the 8Y8M_A strain. This SADS-CoV clade clusters closely with bat coronavirus HKU2, and both are members of the Rhinacovirus subgenus (Figure 8A). By March 2026, a comprehensive metadata analysis was conducted on all available SADS-CoV nucleocapsid protein (N) sequences retrieved from the NCBI database. Among the historical host reservoir of SADS-CoV, swine hosts constitute the overwhelming majority. In terms of geographical distribution, the period from 2016 to 2018 saw a high incidence of SADS-CoV outbreaks in Guangdong Province, China (Figures 8B, C). Notably, our detection introduces a rare sequence originating from an environment sample rather than a direct swine host. This finding not only expands the recognized host categories in public databases but also provides critical evidence for the silent persistence or recent undetected circulation of SADS-CoV in environmental samples.
Figure 8
4 Discussion
Previously, metagenomic studies of Wenzhou water bodies have revealed the interaction between microorganisms and antibiotic resistance genes, which greatly enriches our understanding of the water environment in Wenzhou (). Our research will further analyze the diversity of unknown viromes in the water environment, reveal the impact of virus distribution and diversity, and monitor the potential viral risks in the water environment.
In this study, we found a highly diverse virome in the aquatic environment, including viruses that infect algal, plants, and animals. The bacteriophages of Rudiviridae and Fiersviridae in the lake and Solspiviridae, Kyanoviridae, and Blumeviridae in wastewater were dominant, and this numerical advantage was similar to the research findings of water ecosystems such as the Yangtze River (), rivers () and lake (). A few algal virus families may regulate the population structure of aquatic zooplankton ().
At the same time, we detected the Adenoviridae and the Coronaviridae in both groups of water samples. The Adenoviridae is an established indicator of fecal contamination and an important waterborne pathogen causing human diseases (; ). A major and unexpected finding is the detection of the SADS-CoV in both lake and domestic wastewater samples. To our knowledge, this study reports the detection of SADS-CoV-like sequences in an environmental water sample, as previous studies have strictly isolated this virus from pig farms (Zhou et al., 2018). Interestingly, our sampling sites were located near residential areas and food stalls, so it is possible that the SADS-CoV-like detected in our samples did not originate from live pigs, but more likely from domestic sewage and kitchen waste. Food stalls and residents frequently wash raw pork products. If these pork products are contaminated with the virus, the wash water and food residues can easily introduce viral fragments into the local drainage system. Furthermore, during heavy rainfall, untreated wastewater enters lakes via storm drains, thereby entering the same water system. Previous study suggested that pig feces or feces from other mammals may excrete a large number of viruses, which enter the water environment through runoff or wastewater discharge, which is consistent with the known fecal-oral transmission route (). Some studies have also shown that related coronaviruses can persist under low temperature conditions (). This study provides a basis for early monitoring and preventing the spread of coronavirus in low-temperature water environments.
Although the detection of SADS-CoV-like sequences is noteworthy, the short fragments recovered may be related to the recovery efficiency of the concentration method employed, which warrants careful consideration. The anionic membrane adsorption method relies on electrostatic interactions between viral particles and the negatively charged membrane. Non-enveloped viruses possess a stable protein capsid that is relatively resistant to pH changes and physical forces during concentration, and may therefore be recovered at higher efficiencies. In contrast, enveloped viruses including SADS-CoV have a lipid bilayer that is more vulnerable to the chemical and physical conditions encountered during processing, which may reduce their recovery. The differential recovery of enveloped and non-enveloped viruses has also been observed in previous comparisons of virus concentration methods (; ). Consequently, this may have influenced the observed proportions of different virus types in our virome data. Future studies employing complementary concentration techniques would help address this bias.
Baculoviridae mainly use insects as hosts (). In the research results, the relative abundance of the Baculoviridae-Baculovirus was the highest in both groups of water samples (reaching over 65%), indicating that baculoviruses can continue to exist in an environment without insect hosts, and can stably exist in surface water and sediments for weeks to months (; ).
In addition to the animal virus, we also discovered two specific viruses: Ginkgo biloba picorna-like virus (GBPV) and Ginkgo biloba dicistrovirus (GBDV). These viruses were originally discovered in terrestrial Ginkgo plant leaves (), and our results showed that both viruses belong to the Picornavirales through virome and phylogenetic analysis. The viral sequence and phylogenetic analysis of Ginkgo biloba dicistrovirus belonged to the Dicistroviridae, whose members are almost all insect viruses. The robust, non-enveloped capsid of Dicistroviridae suggested that they are stable in the environment for long periods, allowing them to be transported and detected (). Meanwhile, the phylogenetic analysis showed that the two viruses clustered in the same branch with some arthropods. Secondly, the host prediction results also emphasized this conclusion. It showed that arthropods were the most likely hosts of the two viruses. Ginkgo trees may have served as passive virus reservoirs during the feeding activities of some arthropods. The detection of these viruses in the aquatic environment may be since rainstorms washing the virus particles from the source of urban trees into the water.
Monitoring of the aquatic viral community provides an important perspective for assessing viral contamination in lake and wastewater. Non-etheless, our study still had limitations. First, our sample size is relatively small, which may not fully represent the viral diversity of the whole region. Consequently, the comparative findings presented here should be viewed as preliminary observations. A larger-scale study with additional samples from multiple sites or time points might reveal greater viral diversity and could potentially lead to different comparative outcomes, particularly regarding the relative abundance and presence of specific viral taxa. Second, the viral sequences successfully validated in this study do not confirm the presence of infectious viral particles. Viral viability and infectivity were not assessed in this study. Future studies incorporating cell-culture-based infectivity assays are needed to evaluate whether the viruses detected in these environmental samples have pathogenic potential.Future work should: (i)expand the sample size and sampling scope to further validate our observations; (ii) include negative and process controls in the experimental workflow to distinguish genuine viral signals from potential contamination. (iii) employ deeper sequencing, targeted enrichment, or long-read sequencing technologies to obtain more complete viral genomes. (iv) incorporate cell-culture-based infectivity assays to assess whether the detected viruses have pathogenic potential.
5 Conclusions
In summary, our study showed that aquatic viral communities were complex assemblages influenced by both autochthonous and terrestrial inputs. The eventual detection of Ginkgo biloba picorna-like virus and Ginkgo biloba dicistrovirus highlighted the role of hydrological pathways in the transmission of viral genetic material across ecosystem boundaries. Furthermore, our investigation contributes to characterizing the environmental occurrence and potential public health implications of known coronavirus viruses (e.g., SADS-CoV) in waters environmental. This study expanded our knowledge of the aquatic virosphere and refreshed the understanding of the correlation between viruses in terrestrial and aquatic ecosystems.
Statements
Data availability statement
The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found in the article/Supplementary material.
Author contributions
FX: Writing – original draft, Data curation, Software, Writing – review & editing, Conceptualization. HQ: Visualization, Software, Writing – review & editing. FY: Writing – review & editing, Visualization, Software. ZQ: Visualization, Writing – review & editing, Software. DZ: Writing – review & editing, Visualization, Supervision. NL: Writing – review & editing, Supervision, Validation. PX: Writing – review & editing, Supervision, Validation.
Funding
The author(s) declared that financial support was not received for this work and/or its publication.
Conflict of interest
The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Generative AI statement
The author(s) declared that generative AI was used in the creation of this manuscript. Generative AI (Gemini 3) was used exclusively to assist in checking English grammar and improving language clarity.
Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.
Publisher’s note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
Supplementary material
The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fmicb.2026.1831710/full#supplementary-material
References
1
AhmedW. P. M.BertschA.BivinsK.BibbyK.FarkasA.GathercoleE.et al. (2020). Comparison of virus concentration methods for the RT-qPCR-based recovery of murine hepatitis virus, a surrogate for SARS-CoV-2 from untreated wastewater. Sci Total Environ. 739:139960. doi: 10.1016/j.scitotenv.2020.139960
2
BabayanS. A.OrtonR. J.StreickerD,. G. (2018). Predicting reservoir hosts and arthropod vectors from evolutionary signatures in RNA virus genomes. Science 362,577–580. doi: 10.1126/science.aap9072
3
BiggsT. E. G.Huisma nJ.BrussaardC. P. D. (2021). Viral lysis modifies seasonal phytoplankton dynamics and carbon flow in the Southern Ocean. Isme j.15,3615–3622. doi: 10.1038/s41396-021-01033-6
4
BivinsA.GreavesJ.FischerR. K. C.YindaW. , Ahmed, M.KitajimaV. J.et al. (2020). Persistence of SARS-CoV-2 in water and wastewater. Environ Sci Technol Lett. 7,937–942. doi: 10.1021/acs.estlett.0c00730
5
ButinaT. V.BukinY. S.PetrushinI. S.TupikinA. E.KabilovM. R.BelikovS. I. (2021). Extended evaluation of viral diversity in lake baikal through metagenomics. Microorganisms9:760. doi: 10.3390/microorganisms9040760
6
ChenC.WuY.LiJ.WangX.Z engZ.XuJ.et al. (2023). Xia. TBtools-ii: a “one for all, all for one” bioinformatics platform for biological big-data mining. Mol Plant. 16, 1733–1742. doi: 10.1016/j.molp.2023.09.010
7
ChenG.JiangJ.SunY. (2024). RNAVirHost: a machine learning-based method for predicting hosts of RNA viruses through viral genomes. Gigascience13:giae059. doi: 10.1093/gigascience/giae059
8
ChenS.ZhouY.ChenY.GuJ. (2018). Fastp: an ultra-fast all-in-one FASTQ preprocessor. Bioinformatics34,i884–i890. doi: 10.1093/bioinformatics/bty560
9
ChengR. L.LiX. F.ZhangC. X. (2021). Novel dicistroviruses in an unexpected wide range of invertebrates. Food Environ Virol. 13,423–431. doi: 10.1007/s12560-021-09472-2
10
Cuevas-FerrandoE.Pérez-CataluñaA.AllendeA.GuixS.RandazzoW.SánchezG. (2021). Recovering coronavirus from large volumes of water. Sci Total Environ. 762:143101. doi: 10.1016/j.scitotenv.2020.143101
11
DuK.YangF.ZhangJ. T.YuR. C.DengZ.LiW. F.et al. (2022). Comparative genomic analysis of five freshwater cyanophages and reference-guided metagenomic data mining. Microbiome10:128. doi: 10.1186/s40168-022-01324-w
12
ElfellakiN.BerrouchS.BiaryA.GoïtaS.R afiH.LachkarH.et al. (2024). Comparison of four concentration methods of adenovirus, norovirus and rotavirus in tap water. J. Virol. Methods330:115013. doi: 10.1016/j.jviromet.2024.115013
13
FarkasK.PellettC.Alex-SandersN.BridgmanM. T. P.Cor bishleyA.GrimsleyJ. M. S.et al. (2022). Comparative assessment of filtration- and precipitation-based methods for the concentration of SARS-CoV-2 and other viruses from wastewater. Microbiol. Spectr. 10:e0110222. doi: 10.1128/spectrum.01102-22
14
HammerØ.HarperD. A. T.RyanP. D. (2001). Past: paleontological statistics software package for education and data analysis. Palaeontologia Electronica4, 1–9. Available online at: https://www.semanticscholar.org/paper/PAST%3A-paleontological-statistics-software-package-Hammer-Harper/a82d7e0dc7b2d3a1b159e0902bb9f3017788a786
15
HussainS.WangX.PanC.ChenS.XieJ.MahtabN.et al. (2025). Salinity-driven shifts in estuarine viral community composition and diversity near the Shenzhen coast. Appl Environ Microbiol. 91:e0040725. doi: 10.1128/aem.00407-25
16
KatohK.RozewickiJ.YamadaK. D. (2019). MAFFT online service: multiple sequence alignment, interactive sequence choice and visualization. Brief. Bioinform.20, 1160–1166. doi: 10.1093/bib/bbx108
17
KumthipK.KhamrinP.UshijimaH.ManeekarnN. (2023). Detection of six different human enteric viruses contaminating environmental water in Chiang Mai, Thailand. Microbiol. Spectr. 11:e0351222. doi: 10.1128/spectrum.03512-22
18
LetunicI.BorkP. (2021). Interactive tree of life (iTOL) v5: an online tool for phylogenetic tree display and annotation. Nucleic Acids Res. 49, W293–w296. doi: 10.1093/nar/gkab301
19
LiD.LiuC. M.LuoR.SadakaneK.LamT. W. (2015). MEGAHIT: an ultra-fast single-node solution for large and complex metagenomics assembly via succinct de Bruijn graph. Bioinformatics31, 1674–1676. doi: 10.1093/bioinformatics/btv033
20
LiuC.HuangW.HeX.FengZ.ChenQ. (2024). Research advances on swine acute diarrhea syndrome coronavirus. Animals14:448. doi: 10.3390/ani14030448
21
López-FerberM.CaballeroP.WilliamsT. (2025). Baculovirus genetic diversity and population structure Viruses17:142. doi: 10.3390/v17020142
22
LuG.YeZ. X.QiY. H.LuJ. B.MaoQ. Z.ZhuoJ. C.et al. (2024). Endogenous nege-like viral elements in arthropod genomes reveal virus-host coevolution and ancient history of two plant virus families. J. Virol. 98:e0099724. doi: 10.1128/jvi.00997-24
23
LuJ.RinconN.WoodD. E.BreitwieserF. P.PockrandtC.LangmeadB. (2022a). Metagenome analysis using the Kraken software suite. Nat Protoc. 17, 2815–2839. doi: 10.1038/s41596-022-00738-y
24
LuJ.YangS.ZhangX.TangX.ZhangJ.WangX. (2022b). Metagenomic analysis of viral community in the Yangtze River expands known eukaryotic and prokaryotic virus diversity in freshwater. Virol. Sin. 37, 60–69. doi: 10.1016/j.virs.2022.01.003
25
MinigulovN.BoranbayevK.BekbossynovaA.GadilgereyevaB.FilchakovaO. (2024). Structural proteins of human coronaviruses: what makes them different?Front. Cell. Infect. Microbiol. 14:1458383. doi: 10.3389/fcimb.2024.1458383
26
MiuraT.TakinoH.GimaA.HaramotoE.AkibaM. (2021). Recovery of nucleic acids of enteric viruses and host-specific bacteroidales from groundwater by using an adsorption-direct extraction method. Appl. Environ. Microbiol. 87:e0071021. doi: 10.1128/AEM.00710-21
27
MohanS. V.HemalathaM.KopperiH.RanjithI.KumarA. K. (2021). SARS-CoV-2 in environmental perspective: occurrence. persistence, surveillance, inactivation and challenges. Chem. Eng. J. 405:126893. doi: 10.1016/j.cej.2020.126893
28
NguyenL. T.SchmidtH. A.von HaeselerA.MinhB. Q. (2015). IQ-TREE: a fast and effective stochastic algorithm for estimating maximum-likelihood phylogenies. Mol. Biol. Evol. 32, 268–274. doi: 10.1093/molbev/msu300
29
OwliaeeI.KhaledianM.MahmoudvandS.AminiR.AbneyS. E.BeikpourF. (2024). Global investigation of the presence of adenovirus in different types of water resources: a systematic review. Virusdisease35, 55–65. doi: 10.1007/s13337-023-00857-4
30
Pedrosa de MacenaL. D. G.Castiglia FeitosaR.Couto da SilvaJ.FerreiraF. C.MaranhãoA. G.BrandãoM. L. L.et al. (2023). Environmental assessment of sewage contamination in the surroundings of a marine outfall combining human mastadenovirus and fecal indicator bacteria. Mar. Pollut. Bull. 193:115110. doi: 10.1016/j.marpolbul.2023.115110
31
Rodríguez-HernándezA. P.Martínez-FloresD.Cruz-ReséndizA.Padilla-FloresT.González-FloresR.EstradaK.et al. (2023). Baculovirus display of peptides and proteins for medical applications. Viruses15:411. doi: 10.3390/v15020411
32
TaoZ.ChenP.CuiN.LinX.JiF.LiuY.et al. (2020). Detection of enteroviruses in urban sewage by next generation sequencing and its application in environmental surveillance. Sci. Total Environ. 728:138818. doi: 10.1016/j.scitotenv.2020.138818
33
TianF.WainainaJ. M.Howard-VaronaC.Domínguez-HuertaG.BolducB.GazitúaM. C.et al. (2024). Prokaryotic-virus-encoded auxiliary metabolic genes throughout the global oceans. Microbiome12:159. doi: 10.1186/s40168-024-01876-z
34
WilliamsT.VirtoC.MurilloR.CaballeroP. (2017). Covert infection of insects by baculoviruses. Front Microbiol. 8:1337. doi: 10.3389/fmicb.2017.01337
35
WiltonR.SzalayA. S. (2022). Performance optimization in DNA short-read alignment. Bioinformatics 38, 2081–2087. doi: 10.1093/bioinformatics/btac066
36
XuM.ChengR.YanR.AiH.QiaoS.LiP.et al. (2026). Comprehensive virome analysis of soybean agroecosystems in China reveals geographic patterns and interfacial virus spillovers. Sci China Life Sci. 69, 307–310. doi: 10.1007/s11427-024-3038-x
37
YanQ.LiuX.SunY.ZengW.LiY.ZhaoF.et al. (2022). Swine enteric coronavirus: diverse pathogen-host interactions. Int J Mol Sci. 23:3953. doi: 10.3390/ijms23073953
38
YangS.MaoQ.WangY.HeJ.YangJ.ChenX. (2022). Expanding known viral diversity in plants: virome of 161 species alongside an ancient canal. Environ. Microbiome17:58. doi: 10.1186/s40793-022-00453-x
39
YeS.LiS.SuC.ShiZ.LiH.HongJ.et al. (2023).Characterization of microbial community and antibiotic resistome in intra urban water, Wenzhou China. Front. Microbiol. 14:1169476. doi: 10.3389/fmicb.2023.1169476
40
ZellR.GrothM.SelinkaL.SelinkaH. C. (2024). Diversity of picorna-like viruses in the teltow canal, Berlin, Germany. Viruses16:1020. doi: 10.3390/v16071020
41
ZhangJ.DingW.ZouG.WangX.ZhaoM.GuoS. (2023). Urban pipeline rainwater runoff is an important pathway for land-based microplastics transport to inland surface water: a case study in Beijing. Sci. Total Environ. 861:160619. doi: 10.1016/j.scitotenv.2022.160619
42
ZhangM.Altan-BonnetN.ShenY.ShuaiD. (2022). Waterborne human pathogenic viruses in complex microbial communities: environmental implication on virus infectivity, persistence, and disinfection. Environ. Sci. Technol.56, 5381–5389. doi: 10.1021/acs.est.2c00233
43
ZhangT.YaoJ.YangZ.WangJ.YangK.YaoL. (2024). Re-emergence of severe acute diarrhea syndrome coronavirus (SADS-CoV) in Henan, central China. Vet. Microbiol. 292:110049. doi: 10.1016/j.vetmic.2024.110049
44
ZhangY.YeZ. X.FengX. X.XuZ. T.ChenJ. P.ZhangC. X.et al. (2023). Prevalence of reversed genome organizations for viruses in the family iflaviridae, order picornavirales. Microbiol Spectr. 11:e0473822. doi: 10.1128/spectrum.04738-22
45
ZhouP.FanH.LanT.YangX. L.ShiW. F.ZhangW.et al. (2018). Fatal swine acute diarrhoea syndrome caused by an HKU2-related coronavirus of bat origin. Nature556, 255–258. doi: 10.1038/s41586-018-0010-9
46
ZhouZ. J.QiuY.PuY.HuangX.GeX. Y. (2020). Bioaider: an efficient tool for viral genome analysis and its application in tracing SARS-CoV-2transmission. Sustain. Cities Soc.63:102466. doi: 10.1016/j.scs.2020.102466
Summary
Keywords
domestic wastewater, ecosystem virome, lake, phylogenetic analysis, swine acute diarrhea syndrome coronavirus
Citation
Xu F, Qin H, Yu F, Qiu Z, Zhang D, Li N and Xiao P (2026) Comparative virome analysis of lake and domestic wastewater revealed the unexpected presence of swine acute diarrhea syndrome coronavirus and Ginkgo-associated viruses. Front. Microbiol. 17:1831710. doi: 10.3389/fmicb.2026.1831710
Received
16 March 2026
Revised
12 May 2026
Accepted
19 May 2026
Published
05 June 2026
Volume
17 - 2026
Edited by
Danny Ionescu, Technical University of Berlin, Germany
Reviewed by
Kingsley Chukwudozie, University of Nigeria, Nsukka, Nigeria
Novo, SPC, Oswaldo Cruz Foundation (Fiocruz), Brazil
Updates
Copyright
© 2026 Xu, Qin, Yu, Qiu, Zhang, Li and Xiao.
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
*Correspondence: Nan Li, 20190283@wzu.edu.cn; Pengpeng Xiao, 20190002@wzu.edu.cn
Disclaimer
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.