Grass Carp Reovirus Induces Formation of Lipid Droplets as Sites for Its Replication and Assembly

Abstract

Grass carp is an important commercial fish in China that is plagued by various diseases, especially the hemorrhagic disease induced by grass carp reovirus (GCRV). Nevertheless, the mechanism by which GCRV hijacks the host metabolism to complete its life cycle is unclear. In this study, we performed lipidomic analysis of grass carp liver samples collected before and after GCRV infection. GCRV infection altered host lipid metabolism and increased de novo fatty acid synthesis. Increased de novo fatty acid synthesis induced accumulation of lipid droplets (LDs). LDs are associated with GCRV viroplasms, as well as viral proteins and double-stranded RNA. Pharmacological inhibition of LD formation led to the disappearance of viroplasms, accompanied by decreased viral replication capacity. Moreover, transmission electron microscopy revealed LDs in close association with the viroplasms and mounted GCRV particles. Collectively, these data suggest that LDs are essential for viroplasm formation and are sites for GCRV replication and assembly. Our results revealed the detailed molecular events of GCRV hijacking host lipid metabolism to benefit its replication and assembly, which may provide new perspective for the prevention and control of GCRV. IMPORTANCE Grass carp reovirus (GCRV) is the most virulent pathogen in the genus Aquareovirus, which belongs to the family Reoviridae. GCRV-induced hemorrhagic disease is a major threat to the grass carp aquaculture industry. Viruses are obligate intracellular parasites that require host cell machinery to complete their life cycle; the mechanism by which GCRV hijacks the host metabolism to benefit viral replication and assembly remains unclear. Our study demonstrated that GCRV infection alters host lipid metabolism and increases de novo fatty acid synthesis. The increased de novo fatty acid synthesis induced accumulation of LDs, which act as sites or scaffolds for GCRV replication and assembly. Our findings illustrate a typical example of how the virus hijacks cellular organelles for replication and assembly and hence may provide new insights for the prevention and control of GCRV.

Keywords: assembly; grass carp reovirus; lipid droplets; lipidomic analysis; replication; viroplasms.

FIG 1

GCRV infection altered host lipid metabolism. (A) Major classes of lipids that identified in all the samples by lipidome analysis. (B) Top 20 subclasses of all identified lipids. (C) PCA score plots for the samples from different groups. (D) Numbers of upregulated and downregulated DELs at different time points after GCRV infection. (E) Expression patterns of lipids belonging to GLs. The red dots represent upregulated DELs, while the blue dots represent downregulated DELs.

FIG 2

Fatty acid synthesis is essential for GCRV replication. (A and D) Relative gene expression level of NS80 in cells treated with different concentrations of TOFA or palmitic acid (PA). (B and E) Plaque assay of cells treated with different concentrations of TOFA or PA. (C and F) Virus titers in cells treated with different concentrations of TOFA or PA. ns, no significant difference.

FIG 3

Triglycerides enhanced GCRV replication. (A) Heat map of the top 10 upregulated and downregulated DELs at different time points after GCRV infection. (B and E) Relative gene expression level of NS80 in cells treated with different concentrations of atglistatin or CAY10499. (C and F) Plaque assay of cells treated with different concentrations of atglistatin or CAY10499. (D and G) Virus titers in cells treated with different concentrations of atglistatin or CAY10499.

FIG 4

GCRV infection induced formation of lipid droplets. (A) Oil red O staining of liver samples collected from GCRV-infected and uninfected fish. Scale bar = 50 μm. (B) Calculated oil red O staining areas in GCRV-infected and uninfected samples. (C) Bodipy 493/503 staining of mock-infected or GCRV-infected cells. Scale bar = 10 μm. (D) Calculated LD numbers/cell in mock-infected or GCRV-infected cells.

FIG 5

Lipid droplets associate with GCRV viroplasms. (A and B) Confocal microscopy analysis of the relationship between LDs and viroplasms stained with anti-NS80 (A) or anti-VP5 (B). Cells were mock infected or infected with GCRV, and then LDs and viroplasms were stained with Bodipy 493/503 and anti-NS80 antibody (A) and anti-VP5 (B) antibody, respectively. Scale bar = 10 μm. (C) The precise relationship between LDs and viroplasms. Scale bar = 10 μm.

FIG 6

The relationship between LDs and viroplasms at different time points after GCRV infection. GCO cells were infected with GCRV and harvested at 1, 2, 4, 6, 12, and 24 hpi. Viroplasms and LDs were stained with anti-NS80 antibody and Bodipy 493/503, respectively. Scale bar = 10 μm.

FIG 7

Lipid droplets are essential for viroplasm formation and GCRV replication. (A) Viroplasms and LDs in GCRV-infected cells after treatment with PA or TOFA. GCO cells were infected with GCRV and then treated with PA or TOFA. The LDs and viroplasms were stained with Bodipy 493/503 and anti-NS80 antibody, respectively. Scale bar = 10 μm. (B) The viroplasms and LDs in GCRV-infected cells after treatment with triacsin C. GCO cells were infected with GCRV and then treated with triacsin C. The LDs and viroplasms were stained as described above. Scale bar = 10 μm. (C) Calculated LD numbers/cell in GCRV-infected cells after treatment or not with triacsin C. (D) Relative gene expression level of NS80 in GCRV-infected cells after treatment or not with triacsin C. (E) Plaque assay of GCRV-infected cells after treatment or not with triacsin C. (F) Virus titers in GCRV-infected cells after treatment or not with triacsin C.

FIG 8

GCRV-encoded proteins recruited by LDs. (A and B) Confocal microscopy analysis of the relationship between LDs and ectopic expression of NS80-mCherry fusion protein (A) or VP5-mCherry fusion protein (B). Cells were transfected with pmCherry-NS80 plasmid (A) or pmCherry-VP5 plasmid (B) and then infected with GCRV or treated with PA to induce LD formation. The relationship between LDs and NS80-mcherry fusion protein (A) or VP5-mcherry fusion protein (B) was analyzed by confocal microscopy. Scale bar = 10 μm.

FIG 9

LDs are sites for GCRV replication and assembly. (A) The relationship between LDs and viral RNA. GCO cells were mock infected or infected with GCRV, and then LDs and viral RNA were stained with Bodipy 493/503 and a specific antibody for double-stranded RNA (dsRNA), respectively. Scale bar = 10 μm. (B) Relationship between LDs and VP2 in the presence of GCRV. GCO cells were transfected with VP2-mCherry plasmid or empty mCherry plasmid and then infected with GCRV to induce LD formation. Then cells were harvested at 24 h for confocal microscopy observation. Scale bar = 10 μm. (C and D) Transmission electron microscopy (TEM) analysis of mock-infected (C) or GCRV-infected (D) cells. N, nucleus; V, viroplasms; LDs, lipid droplets; GCRV, mounted GCRV particles. Scale bar = 2 μm. (E and F) Magnification of TEM pictures of GCRV-infected cells. Scale bar = 500 nm.