key: cord-0256645-m0e1rtxe
authors: Costa, Vincenzo A.; Mifsud, Jonathon C.O.; Gilligan, Dean; Williamson, Jane E.; Holmes, Edward C.; Geoghegan, Jemma L.
title: Metagenomic sequencing reveals a lack of virus exchange between native and invasive freshwater fish across the Murray-Darling Basin, Australia
date: 2021-02-25
journal: bioRxiv
DOI: 10.1101/2021.02.25.432824
sha: cccecda757d3a1d282ed353b12b859915dc6b465
doc_id: 256645
cord_uid: m0e1rtxe

Biological invasions are among the biggest threats to freshwater biodiversity. This is increasingly relevant in the Murray-Darling Basin, Australia, particularly since the introduction of the common carp (Cyprinus carpio). This invasive species now occupies up to 90% of fish biomass, with hugely detrimental impacts on native fauna and flora. To address the ongoing impacts of carp, cyprinid herpesvirus 3 (CyHV-3) has been proposed as a potentially effective biological control. Crucially, however, it is unknown whether CyHV-3 and other cyprinid herpesviruses already exist in the Murray-Darling. Further, little is known about those viruses that naturally occur in wild freshwater fauna, and the frequency with which these viruses jump species boundaries. To document the evolution and diversity of freshwater fish viromes and better understand the ecological context to the proposed introduction of CyHV-3, we performed a meta-transcriptomic viral survey of invasive and native fish across the Murray-Darling Basin, covering over 2,200 km of the river system. Across a total of 36 RNA libraries representing 10 species, we failed to detect CyHV-3 nor any closely related viruses. Rather, meta-transcriptomic analysis identified 18 vertebrate-associated viruses that could be assigned to the Arenaviridae, Astroviridae, Bornaviridae, Caliciviridae, Coronaviridae, Chuviridae, Flaviviridae, Hantaviridae, Hepeviridae, Paramyxoviridae, Picornaviridae, Poxviridae, Reoviridae and Rhabdoviridae families, and a further 27 that were deemed to be associated with non-vertebrate hosts. Notably, we revealed a marked lack of viruses that are shared among invasive and native fishes sampled here, suggesting that there is little virus transmission from common carp to native fish species. Overall, this study provides the first data on the viruses naturally circulating in a major river system and supports the notion that fish harbour a large diversity of viruses with often deep evolutionary histories. Author Summary The ongoing invasion of the common carp in the Murray-Darling Basin, Australia, has wreaked havoc on native freshwater ecosystems. This has stimulated research into the possible biological control of invasive carp through the deliberate release of the virus cyprinid herpesvirus 3 (CyHV-3). However, little is known on the diversity of viruses that naturally circulate in wild freshwater fauna, whether these viruses are transmitted between invasive and native species, nor if CyHV-3 or other cyprinid herpesviruses are already present in the basin. To address these fundamental questions we employed meta-transcriptomic next-generation sequencing to characterise the total assemblage of viruses (i.e. the viromes) in three invasive and seven native fish species cohabiting at 10 sites across 2,200 km of the river system. From this analysis we identified 18 vertebrate-associated viruses across 14 viral families, yet a marked lack of virus transmission between invasive and native species. Importantly, no CyHV-3 was detected. This study shows that freshwater fish harbour a high diversity and abundance of viruses, that viruses have likely been associated with fish for millennia, and that there is likely little direct virus transmission between introduced and native species.

Anthropogenic stressors such as pollution, climate change and the introduction of exotic 62 species continue to pose a significant threat to freshwater habitats, with almost one third of 63 all fish species threatened by extinction [1] . The Murray-Darling Basin, the largest 64 freshwater river system in Australia, harbours at least 12 exotic freshwater fish species [2] . isolated from farmed carp in the late 1990s [9] . Since its discovery, it has been responsible 80 for large disease outbreaks worldwide with a mortality rate of up to 80% in domestic carp 81 [10]. CyHV-3 is transmitted horizontally through direct contact with skin lesions or secretion 82 of viral particles in freshwater where it can survive for up to three days [11] . The host range 83 of CyHV-3 is currently limited to koi and common carp [9] . While CyHV-3 DNA has been 84 identified in goldfish (Carrasius auratus) [12] , it is still relatively unclear whether infection 85 occurs in these species [14, 15] . 86 Initial laboratory trials suggest that CyHV-3 is safe for non-target species [7] . However, little 87 is known about the viruses that naturally circulate in Australian native freshwater fauna, In addition, despite representing a small fraction of the earth's surface water, freshwater 104 environments serve as a habitat for 40-50% of total fish species, harbouring the greatest 105 biodiversity per land area [20] . Such habitats are subject to rapid environmental change, 106 which may significantly impact species connectivity [21] . Since contact and exposure 107 between hosts are vital for cross-species transmission of viruses [22] , these species may 108 also inform us on the ecological factors that impact virome composition within a given host. Tissue specimens (liver and gills) were placed in RNALater and stored in a portable -80°C 136 freezer, then later in a -80°C freezer in the laboratory at Macquarie University, Sydney. for one minute at 5000 rpm. To further homogenise tissue samples and remove tissue 152 residues, the homogenate was centrifuged at full speed for three minutes. The homogenate 153 was carefully removed and RNA from the clear supernatant was extracted using the 154 RNeasy Plus Mini Kit (Qiagen, Hilden, Germany) following the manufacturer's protocol.

Extracted RNA was quantified using NanoDrop (ThermoFisher) and RNA from each species 156 was pooled corresponding to the site in which they were captured, resulting in 36 sample 157 libraries (SI Table 2 sequencing library. We also used this approach to estimate the relative abundance of a 204 host reference gene, ribosomal protein S13 (RPS13), that is stably expressed in fish. To 205 assess any differences in virome composition between hosts and sites, we calculated alpha 206 diversity (virome richness and Shannon diversity) using Rhea packages [28] . Generalised 207 linear models (GLM) were used to identify the impact of host taxonomy (i.e. species), host 208 geography (i.e. site), water temperature, water pH, water turbidity and species origin (i.e.

invasive or native) on both vertebrate-associated virus composition (abundance, richness 210 and diversity) as well as those viruses likely associated with non-fish hosts: the latter should 211 not be affected by aspects of fish biology and hence effectively constitute a negative 212 control. All GLM models were tested using a likelihood-ratio test (χ2) and a Tukey's post 213 hoc analysis (glht) was performed using the multcomp package [29] . To assess viral 214 diversity between samples, we calculated beta diversity using a Bray Curtis dissimilarity 

We characterised the viromes of ten freshwater ray-finned fish species across seven Abundance and diversity of viruses 237 We identified 18 viral sequences that were associated with vertebrate hosts and a further 238 27 that were likely associated with algae, invertebrates and protists in the freshwater 239 environment (SI Figures 1 and 2 ). Because such non-vertebrate viruses were likely derived 240 from diet or contamination of gill tissue, we primarily focused on vertebrate-associated 278 We now describe each of these groups in turn. 280 We identified -ssRNA viruses that occupied phylogenetic positions that were broadly 281 indicative of long-term virus-host co-divergence, with many fish viruses falling basal to 282 reptile and mammalian viruses ( Figure 4) Positive-sense single-stranded (+ssRNA) viruses 346 We identified a viral sequence in common carp that shared 50.7% RdRp sequence Aquareovirus that are known to cause considerable disease in some fish species [42] 408 A key observation of our study was the absence of Cyprinid herpesviruses, including CyHV- Table 1 ). Both SGPV and 417 western carp-gudgeon poxvirus form a highly divergent clade within the subfamily 418 Chordopoxvirinae that is strongly indicative of virus-host co-divergence (Figure 9 ). associated hepeviruses such that their true hosts could not be easily determined. 472 We also examined whether there were any differences in alpha and beta diversity between 473 native and invasive freshwater fishes. Accordingly, we found no association between host 474 origin (i.e. invasive or native) and virome abundance (p = 0.390). When assessing alpha 475 diversity, we similarly found no association between host origin and virome richness (p = 476 0.626) nor Shannon diversity (p = 0.425). Likewise, this result was also observed when 477 examining beta diversity (p = 0.602). 479 To assess any associations between host ecology and non-vertebrate viruses, we similarly 480 performed GLMs using the aforementioned ecological factors as a negative control. As 

The only instance of co-occurrence of viruses from the same family in both invasive and 506 native species were the presence of arenaviruses in native carp-gudgeon and invasive 507 mosquitofish. However, these viruses were so divergent that they likely represent ancient 508 common ancestry rather than recent cross-species transmission (Figure 4) . Eastern ]. In addition, it has been widely suggested that introduced populations are associated 599 with a lower pathogen prevalence and diversity than native species [73, 74] . For example, 600 because invasive species are often established from a small founder population, they likely 601 acquire only a small proportion of pathogens in the novel environment [73, 74] . Once a 602 species rapidly becomes invasive, the diversity of pathogens in this population should 603 remain small, such that the lack of disease likely facilitates the success of invasive species 604 [73, 74] . 605 It is important to note, however, that there were necessary variations within our sampling.

For instance, carp and native fish species were sampled together at 10 out of 13 sites, with 607 carp sampled from all 13 sites ( Figure 1 ). In addition, all other fish species were sampled 608 from 1-5 sites. While an artifact of the distribution of the fishes, such gaps limit the power of 

The world's forgotten fishes

Management of alien fishes in the Murray-Darling 654

Cyprinus carpio) as a powerful invader in Australian waterways

Population dynamics of invading freshwater 658 fish: common carp (Cyprinus carpio) in the Murray-Darling Basin

Ecological effects of common 661 carp (Cyprinus carpio) in a semi-arid floodplain wetland

Meeting the challenge of 664 quantitative risk assessment for genetic control techniques: a framework and some 665 methods applied to the common Carp (Cyprinus carpio) in Australia

Cyprinid herpesvirus 3 668 as a potential biological control agent for carp (Cyprinus carpio) in Australia: susceptibility 669 of non-target species

Viral biocontrol of invasive vertebrates: lessons from 671 the past applied to cyprinid herpesvirus-3 and carp (Cyprinus carpio) control in Australia

Detection and 674 significance of koi herpesvirus (KHV) in freshwater environments

Survival of koi herpesvirus (KHV) in 679 environmental water

Susceptibility of koi × crucian carp and koi × goldfish hybrids to koi herpesvirus (KHV) and 682 the development of KHV disease (KHVD)

Australia: can a carp herpesvirus (CyHV-3) deliver safe and effective ecological restoration? 685

The outbreak of carp disease 687 caused by CyHV-2 as a model for new emerging viral diseases in aquaculture: a review

Pimephales promelas and Carassius auratus to a strain of koi herpesvirus

Meta-transcriptomics and the evolutionary biology of 693 RNA viruses

Using metagenomics to characterize an expanding 695 virosphere

The evolutionary history of 697 vertebrate RNA viruses

The diversity, evolution and origins of 699 vertebrate RNA viruses

Freshwater biodiversity: importance, threats, status and conservation challenges

Ecology and emergence of diseases in fresh waters

Cross-706 species virus transmission and the emergence of new epidemic diseases

Trimmomatic: a flexible trimmer for Illumina sequence 709 data

De novo 711 transcript sequence reconstruction from RNA-seq using the Trinity platform for reference 712 generation and analysis

Fast and sensitive protein alignment using DIAMOND

Geneious 716 basic: an integrated and extendable desktop software platform for the organization and 717 analysis of sequence data

MAFFT Multiple Sequence Alignment Software Version 7: 719 improvements in performance and usability

Rhea: a transparent and modular R 721 pipeline for microbial profiling based on 16S rRNA gene amplicons

Simultaneous inference in general parametric models

An R Package for reproducible interactive analysis 726 and graphics of microbiome census data

VEGAN, a package of R functions for community ecology

trimAl: a tool for automated 731 alignment trimming in large-scale phylogenetic analyses

IQ-TREE: a fast and effective 734 stochastic algorithm for estimating maximum-likelihood phylogenies

ModelFinder: 737 fast model selection for accurate phylogenetic estimates

Taxonomy of the order Bunyavirales: update 2019

Virome composition in marine fish revealed by meta-transcriptomics

New virus of the family Flaviviridae 745 detected in lumpfish (Cyclopterus lumpus)

Genome 747 sequence analysis of Tamana bat virus and its relationship with the genus Flavivirus

First isolation of a novel 750 aquatic flavivirus from chinook salmon (Oncorhynchus tshawytscha) and its replication in a 751 piscine animal model

Identification of reptarenaviruses, hartmaniviruses, and a novel chuvirus in captive native 754

Brazilian boa constrictors with boid Inclusion body disease

Transcriptomics sequencing provides 757 insights into understanding the mechanism of grass carp reovirus infection

Completion of the rabies virus 760 genome sequence determination: highly conserved domains among the L (polymerase) 761 proteins of unsegmented negative-strand RNA viruses

Hidden 763 diversity and evolution of viruses in market fish

Redefining the invertebrate 765 RNA virosphere

ICTV Virus Taxonomy Profile: 767 Nodaviridae

Australian freshwater fish with epizootic haematopoietic necrosis virus (EHNV)

Characterization of a novel picornavirus isolate from a diseased European eel (Anguilla 773 anguilla)

ICTV Virus 775 Taxonomy Profile: Caliciviridae

Characterization of a novel calicivirus causing systemic infection in Atlantic salmon (Salmo 778 salar L.): proposal for a new genus of Caliciviridae

Salmon gill 780 poxvirus, the deepest representative of the Chordopoxvirinae

Virome of > 12 783 thousand Culex mosquitoes from throughout California

Complete genome sequence of a novel bastrovirus 785 isolated from raw sewage

Detection of a novel 787 RNA virus with hepatitis E virus-like non-structural genome organization in amphibian, agile 788 frog (Rana dalmatina) tadpoles

Salmon gill poxvirus, a recently 790 characterized infectious agent of multifactorial gill disease in freshwater and seawater-791 reared Atlantic salmon

Isolation of a chinook 793 salmon bafinivirus (CSBV) in imported goldfish L. in the United Kingdom and evaluation of 794 its virulence in resident fish species

Description and initial 796 characterization of metatranscriptomic nidovirus-like genomes from the proposed new 797 family Abyssoviridae

Discovery of a 800 novel bottlenose dolphin coronavirus reveals a distinct species of marine mammal 801 coronavirus in Gammacoronavirus

Identification of a 803 novel coronavirus from a beluga whale by using a panviral microarray

Endangered wild salmon infected by newly discovered viruses. eLife

Low diversity and high levels of population genetic 808 structuring in introduced eastern mosquitofish (Gambusia holbrooki) in the greater 809

Melbourne area

Do invasive 811 eastern gambusia shape wetland fish assemblage structure in south-eastern Australia?

Geographic distribution of 814 (EHNV) in freshwater fish in south eastern Australia: lost opportunity for a notifiable 815 pathogen to expand its geographic range

Unprecedented genomic diversity of RNA viruses in arthropods reveals the ancestry of 818 negative-sense RNA viruses

Genome sequences 820 of five arboviruses in field-captured mosquitoes in a unique rural

The evolution and 823 genetics of virus host shifts

Taxonomy of the 825 order Mononegavirales: update 2019

Comprehensive 827 phylogeny of ray-finned fishes (Actinopterygii) based on transcriptomic and genomic data

Phylogenetic 830 classification of bony fishes

Phylogenetic relationships 832 among rhabdoviruses inferred using the L polymerase gene

Evolutionary ecology of virus emergence

Australia's national list of reportable diseases of 837 aquatic animals

Parasites and 840 pathogens lag behind their host during periods of host range advance

Introduced species and their 843 missing parasites

Haematopoietic necrosis in a goldfish (Carassius 845 auratus) associated with an agent morphologically similar to herpesvirus

Incursions 848 of Cyprinid herpesvirus 2 in goldfish populations in Australia despite quarantine practices

Deciphering the origin and 851 evolution of hepatitis b viruses by means of a family of non-enveloped fish viruses

Distinct viral 854 lineages from fish and amphibians reveal the complex evolutionary history of 855

The molecular epidemiology 857 of iridovirus in Murray cod (Maccullochella peelii peelii) and dwarf gourami (Colisa lalia) from 858 distant biogeographical regions suggests a link between trade in ornamental fish and 859 emerging iridoviral diseases

Australian freshwater fishes to dwarf gourami iridovirus (Infectious spleen and kidney 862 necrosis virus)