Enormous diversity of RNA viruses in economic crustaceans

Abstract

Crustaceans are important food sources worldwide and possess significant ecological status in the marine ecosystem. However, our understanding of the diversity and evolution of RNA viruses in crustaceans, especially in economic crustaceans, is still limited. Here, 106 batches of economic crustaceans including 13 species were collected from 24 locations in China during 2016-2021. We identified 90 RNA viruses, 69 of which were divergent from the known viruses. Viral transcripts were assigned to 18 different viral families/clades and three unclassified groups. Among the identified viruses, five were double-stranded RNA viruses, 74 were positive-sense single-stranded RNA (+ssRNA) viruses, nine were negative-sense single-stranded RNA (-ssRNA) viruses, and two belonged to an unclassified RNA virus group. Phylogenetic analyses showed that crustacean viruses were often clustered with viruses identified from invertebrates. Remarkably, most crustacean viruses were closely related to those from different host species along the same food chain or ecological aquatic niche. In addition, the genome structures of the newly discovered picornaviruses exhibited remarkable diversity. Our study significantly expands the diversity of viruses in important economic crustaceans and provides essential data for the risk assessment of the pathogens spreading in the global aquaculture industry.

Importance: The study delves into the largely uncharted territory of RNA viruses in crustaceans, which are not only vital for global food supply but also play a pivotal role in marine ecosystems. Focusing on economic crustaceans, the research uncovers 90 RNA viruses, with 69 being potentially new to science, highlighting the vast unknown viral diversity within these marine organisms. The findings reveal that these viruses are often related to those found in other invertebrates and tend to share close relationships with viruses from species within the same food web or habitat. This suggests that viruses may move between different marine species more frequently than previously thought. The discovery of such a wide variety of viruses, particularly the diverse genome structures of newly identified picornaviruses, is a significant leap forward in understanding the crustacean virology. This knowledge is crucial for managing disease risks in aquaculture and maintaining the balance of marine ecosystems.

Keywords: crustaceans; diversity; evolution; virome.

Fig 1

The distribution of crustaceans and the viruses identified in this study. (a) Geographic distribution of the crustaceans surveyed between July 2016 and December 2021 (N = 106 libraries). The pie chart showed the number and proportion of the species sampled in the eight provinces and one sea area. Different colors represented different host species. (b) Phylogenetic relationships of the cytochrome C oxidase subunit I (cox1) gene sequences of crustacean species in this study. Different colors indicate different host species. (c) Distribution of the samples by living condition, health condition (middle panel), sampling location (bottom panel), and sampling year. See also Table S1.

Fig 2

Crustacean viruses were analyzed in this study. (a) Distribution and abundance of crustaceans-associated viruses across different hosts. The relative abundance of the viruses in each library was calculated and normalized by the number of FPKM. Viral species from 18 families, 1 unclassified Picornavirales, 1 unclassified Bunyavirales, and 1 unclassified RNA virus were shown, with viral taxonomy (family) and host abundance indicated by different colors. “No reference sequence in GenBank” means that the cox1 reference sequence for the related species has not been made available in the GenBank. (b) Number of virus species in the corresponding libraries (Fig. 2a), with viral taxonomy (family) indicated by different colors.

Fig 3

Maximum likelihood phylogenetic analyses of the RdRp protein sequences of the double-sense RNA viruses identified in crustaceans. Multiple sequence alignment was performed using MAFFT v7.407 with the E-INS-I algorithm. Non-conserved amino acid regions were removed using Trimal v.1.4. We selected the best-fit amino acid substitution model identified using ModelFinder and performed phylogenetic analyses using the maximum likelihood method embedded in IQ-TREE v.2.1.4 with 1,000 bootstrap replicates. Red bold lines indicate the viruses found in this study. Taxon names of the reference sequences downloaded from GenBank within the same cluster with crustacean viruses found in this study were colored by the apparent host group from which the viral sequence originated. Additionally, the reference crustacean viruses within the same cluster as those found in this study were highlighted. The host ecosystem was distinguished by different colored solid circles. All the phylogenetic trees were mid-point rooted for clarity and only bootstrap values ≥60% were shown. The viruses identified in this study were highlighted in red and bold. Best-fit amino acid substitution model: (a) Q.pfam + F + R6; (b) Q.pfam + F + R4.

Fig 4

Maximum likelihood phylogenetic analyses of the RdRp protein sequences of the negative-sense RNA viruses identified in crustaceans. Multiple sequence alignment was performed using MAFFT v.7.407 with the E-INS-I algorithm. Non-conserved amino acid regions were removed using Trimal v.1.4. We selected the best-fit amino acid substitution model identified using ModelFinder and performed phylogenetic analyses using the maximum likelihood method embedded in IQ-TREE v.2.1.4 with 1,000 bootstrap replicates. Red bold lines indicate the viruses found in this study. Taxon names of the reference sequences downloaded from GenBank within the same cluster with crustacean viruses found in this study were colored by the apparent host group from which the viral sequence originated. Additionally, the reference crustacean viruses within the same cluster as those found in this study were highlighted. The host ecosystem was distinguished by different colored solid circles. All the phylogenetic trees were mid-point rooted for clarity and only bootstrap values ≥60% were shown. The viruses identified in this study were highlighted in red and bold. Best-fit amino acid substitution model: (a) LG + I + G4; (b) LG + F + I + G4; (c) Q.pfam + F + R7.

Fig 5

Maximum likelihood phylogenetic analyses of the RdRp protein sequences of the positive-sense RNA viruses identified in crustaceans. Multiple sequence alignment was performed using MAFFT v.7.407 with the E-INS-I algorithm. Non-conserved amino acid regions were removed using Trimal v.1.4. We selected the best-fit amino acid substitution model identified using ModelFinder and performed phylogenetic analyses using the maximum likelihood method embedded in IQ-TREE v.2.1.4 with 1,000 bootstrap replicates. Red bold lines indicate the viruses found in this study. Taxon names of the reference sequences downloaded from GenBank within the same cluster with crustacean viruses found in this study were colored by the apparent host group from which the viral sequence originated. Additionally, the reference crustacean viruses within the same cluster as those found in this study were highlighted. The host ecosystem was distinguished by different colored solid circles. All the phylogenetic trees were mid-point rooted for clarity and only bootstrap values ≥60% were shown. Viruses identified in this study were highlighted in red and bold. Best-fit amino acid substitution model: (a) Q.pfam + F + R5, (b) LG + G4, (c) LG + I + G4, (d) Q.pfam + G4, (e) LG + F + I + G4, (f) Q.pfam + F + I + G4, (g) LG + I + G4, (h) PMB + R5, (i) LG + F + I + G4, (j) Q.pfam + F + R6.

Fig 6

Predicted genome structure of the crustacean viruses. The conserved domains of the viruses were visualized and viruses with segmented genomes were indicated by asterisks. The genomes are depicted at a consistent length scale, as indicated at the bottom of the illustration. Within each genome, the outlined boxes represent the predicted ORF boundaries. Colored boxes highlight the regions that exhibit relatively high BLAST homology to viral proteins/domains, with specifics detailed in the lower left corner of the figure.