Natural and experimental infection of Caenorhabditis nematodes by novel viruses related to nodaviruses - PubMed (original) (raw)

. 2011 Jan 25;9(1):e1000586.

doi: 10.1371/journal.pbio.1000586.

Alyson Ashe, Joséphine Piffaretti, Guang Wu, Isabelle Nuez, Tony Bélicard, Yanfang Jiang, Guoyan Zhao, Carl J Franz, Leonard D Goldstein, Mabel Sanroman, Eric A Miska, David Wang

Affiliations

Marie-Anne Félix et al. PLoS Biol. 2011.

Abstract

An ideal model system to study antiviral immunity and host-pathogen co-evolution would combine a genetically tractable small animal with a virus capable of naturally infecting the host organism. The use of C. elegans as a model to define host-viral interactions has been limited by the lack of viruses known to infect nematodes. From wild isolates of C. elegans and C. briggsae with unusual morphological phenotypes in intestinal cells, we identified two novel RNA viruses distantly related to known nodaviruses, one infecting specifically C. elegans (Orsay virus), the other C. briggsae (Santeuil virus). Bleaching of embryos cured infected cultures demonstrating that the viruses are neither stably integrated in the host genome nor transmitted vertically. 0.2 µm filtrates of the infected cultures could infect cured animals. Infected animals continuously maintained viral infection for 6 mo (∼50 generations), demonstrating that natural cycles of horizontal virus transmission were faithfully recapitulated in laboratory culture. In addition to infecting the natural C. elegans isolate, Orsay virus readily infected laboratory C. elegans mutants defective in RNAi and yielded higher levels of viral RNA and infection symptoms as compared to infection of the corresponding wild-type N2 strain. These results demonstrated a clear role for RNAi in the defense against this virus. Furthermore, different wild C. elegans isolates displayed differential susceptibility to infection by Orsay virus, thereby affording genetic approaches to defining antiviral loci. This discovery establishes a bona fide viral infection system to explore the natural ecology of nematodes, host-pathogen co-evolution, the evolution of small RNA responses, and innate antiviral mechanisms.

PubMed Disclaimer

Conflict of interest statement

Eric Miska's spouse is a member of the PloS Biology editorial staff; in accordance with the PLoS policy on competing interests she has been excluded from all stages of the review process for this article.

Figures

Figure 1

Figure 1. Intestinal cell infection phenotypes in wild Caenorhabditis isolates.

(A–H) C. briggsae JU1264 and (I,J) C. elegans JU1580 observed by Nomarski microscopy. (A–C, E–G, I) Infected adult hermaphrodites from the original cultures, with the diverse infection symptoms: convoluted apical intestinal border (A), degeneration of intestinal cell structures and liquefaction of the cytoplasm (B, G, I), presence of multi-membrane bodies (C). The animals in (E–H) were also observed in the fluorescence microscope after live Hoechst 33342 staining of the nuclei, showing the elongation and degeneration of nuclei (E′–H′). In (E), the nucleus and nucleolus are abnormally elongated. In (F), the nuclear membrane is no longer visible by Nomarski optics. In (G), the cell cytoplasmic structures are highly abnormal (apparent vacuolisation) and the nucleus is very reduced in size. In (E–H′), arrows denote nuclei and arrowheads nucleoli. The infected animal in (I) displays an abnormally large intestinal cell that is probably the result of cell fusions, with degeneration of cellular structures including nuclei. (D, H, J) Uninfected (bleached) adults. Arrowheads in (J) indicate antero-posterior boundaries between intestinal cells, each of which generally contains two nuclei. Bars: 10 µm. (K) Proportion of worms showing the indicated cumulative number of morphological infection symptoms in at least one intestinal cell, in the original wild isolate (I), after bleaching (bl) and after re-infection by a 0.2 µM filtrate (RI). Note that not all symptoms shown in (A–I) were scored, because some are difficult to score or may also occur in healthy animals. The animals were scored 4 d after re-infection for C. briggsae JU1264, and 7 d after re-infection for C. elegans JU1580, at 23°C. The symptoms are similar in both species, and generally more frequent in JU1264. *** p value on number of worms showing infection symptoms <7.10−11, Fisher's exact test; ** p value<3.10−6; * p value<3.10−2.

Figure 2

Figure 2. Transmission electron micrographs of intestinal cells of C. elegans JU1580 adult hermaphrodites.

(A,D,G) Bleached animals. (B–C, E–F, H) Naturally infected animals. (A–C) The infection provokes a reorganization of cytoplasmic structures, most visibly the loss of intestinal lipid storage granules (g). The cytoplasm of infected intestinal cells mostly contains rough endoplasmic reticulum (rer) and mitochondria (m). * hole in the resin used for inclusion in electron microscopy. (D–F) A nucleus in a non-infected animal is surrounded by a nuclear membrane (see inset in D), whereas the nuclear membrane disappears upon infection (E–F). Absence or incomplete nuclear membrane was observed repeatedly in infected animals, while the nuclear membrane could be observed on bleached animals (using both fixation methods). The nuclear material (n) in (F) may represent nucleolar material and at lower magnification (not shown) matches the shape of elongated nucleoli as observed by Nomarski optics (Figure 1E–G). The rough endoplasmic reticulum (rer) on the left of the mitochondrion (m) in (F) may be a remnant of the nuclear envelope. (G–H) The infection may result in disorganization of the intermediate filament (IF) network normally located below the apical plasma membrane. On the right of (H) is shown a higher magnification of the intestinal lumen, showing putative viral particles (arrowheads). The animals were fixed using high-pressure freezing (A–C, E–F) or conventional fixation (D, G, H).

Figure 3

Figure 3. Genomic organization and phylogenetic analysis of novel viruses.

(A) Schematic of genomic organization of Santeuil virus. Predicted open reading frames are displayed in gray boxes. Red bar indicates sequence used to generate double-stranded DNA probes for Northern blotting. Blue bar indicates sequence used to generate single-stranded riboprobes. (B) Neighbor-joining phylogenetic analysis of the predicted RNA-dependent RNA polymerases encoded by the RNA1 segments. (C) Neighbor-joining phylogenetic analysis of the predicted capsid proteins encoded by the RNA2 segments.

Figure 4

Figure 4. Molecular evidence of viral infection.

(A) RT-PCR detection of the Orsay virus in the original JU1580 wild isolate (I), after bleaching (bl) and after re-infection by a 0.2 µM filtrate after 7 d (RI1) and 3 wk (RI2) of culture at 23°C. (B) RT-PCR detection of the Santeuil virus in the original wild isolate (I), after bleaching (bl) and after re-infection by a 0.2 µM filtrate after 4 d (RI1) and 4 wk (RI2) at 23°C. (C) Northern blots of Santeuil virus RNA1 and RNA2 segments hybridized using a double-stranded DNA probe. (D) Northern blots of Santeuil virus RNA1 segment using + and − sense riboprobes. (E) RNA FISH with a probe targeting Orsay virus RNA1 segment. Representative JU1580bl animals following infection by Orsay virus (top and middle rows) or uninfected (bottom row). S corresponds to ovary sheath cells, OO is an oocyte, and I is an intestinal cell.

Figure 5

Figure 5. Specificity of infection by the Orsay and Santeuil viruses.

(A) Specificity of infection by the Orsay virus. Each Caenorhabditis strain (name indicated below the gel) was mock-infected (−) or infected with a virus filtrate (+). RT-PCR on cultures after 7 d at 23°C. See Figure S3 for corresponding morphological symptom scoring. (B) Specificity of infection by the Santeuil virus. RT-PCR results after 4 d at 23°C. (C) Quantitative variation in viral replication N2 versus JU1580. N2 and JU1580 were tested by qRT-PCR for infection with Orsay virus extract (n = 10 independent replicates for each strain). By conventional RT-PCR assay, Orsay virus infection of N2 yielded positive bands in 3 out of 10 replicate infections whereas 7 out of 10 replicate infections of JU1580bl were positive in these conditions. Control RNA (n = 6) was extracted from JU1580bl animals grown in parallel without virus filtrate, and to which filtrate was added at the time of sample collection. RNA levels were normalized to ama-1 and shown as average fold-change relative to JU1580bl. Error bars represent SEM.

Figure 6

Figure 6. Small RNAs produced upon viral infection.

Number of unique sequences obtained by Illumina/Solexa high-throughput sequencing of a 5′-independent small RNA library from JU1580 matching a given position in the Orsay virus segment RNA1 (A) or RNA2 (B). The number of sequences in sense and antisense orientation are shown on the positive (blue) and negative (red) _y_-axis, respectively. Only sequences with a perfect and unambiguous match to the virus genome were considered. The location of virus protein-coding genes is indicated below each graph as black bars and the RNA genome as a line. Features of sense and antisense sequences (length and identity of first nucleotide) are shown to the right of each graph.

Figure 7

Figure 7. RNAi-deficient mutants of C. elegans can be infected by the Orsay virus.

(A) JU1580bl, N2, rde-1(ne219) (n = 10 independent replicates each), rde-2(ne221), rde-4(ne301), and mut-7(pk204) (n = 5 independent replicates each) were tested by qRT-PCR for infection with Orsay virus extract. RNA levels were normalized to ama-1 and shown as average fold-change relative to JU1580bl. Error bars represent SEM. Same results as displayed in Figure 5C for N2 and JU1580. (B) Scoring of symptoms in two independent replicates of infection of rde-1 mutant and wild-type N2 animals by the Orsay virus filtrate, after 4 d.

Figure 8

Figure 8. Natural variation in somatic RNAi efficacy in C. elegans.

(A) Somatic RNAi was tested using bacteria expressing dsRNA specific for the unc-22 gene (acting in muscle; [37]). The percentage of animals with the corresponding twitcher phenotype is shown for different C. elegans wild isolates (representative of the species' diversity; [38]). Bar: standard error over four replicate plates. (B) Germline RNAi was tested by feeding the animals with bacteria expressing dsRNA specific for the pos-1 gene. The percentage of animals with the corresponding embryonic-lethal phenotype is shown for five wild genetic backgrounds of C. elegans. Cbr-lin-12 RNAi is a negative control. Bar: standard error over six replicate plates (too small to be seen). _n_>450 observed individuals for each treatment. (C) Somatic RNAi was tested using bacteria expressing dsRNA specific for GFP. Each point corresponds to the median log2(GFP/DsRed) intensity ratio from one flow cytometry run of strains carrying the let-858::GFP transgene in the JU1580 and N2 backgrounds, after treatment with GFP RNAi or empty vector. Horizontal bars indicate group means. The difference in log2 intensity ratios between GFP RNAi and empty vector is reduced in JU1580 compared to N2 (p<0.001, see Methods). (D) unc-22 dsRNA was administered by injection into the syncytial germline of the mother. 10–14 animals of each genotype were injected and 30 progeny were scored for the twitcher phenotype on each plate. (E) Orsay virus sensitivity of seven wild C. elegans isolates representative of the species' diversity. Morphological symptoms were scored 5 d after infection of clean cultures by the Orsay virus filtrate at 23°C. The JU1580 control was performed in duplicate. Bar: standard error on total proportion. *** p<0.001.

Comment in

References

    1. Hao L, Sakurai A, Watanabe T, Sorensen E, Nidom C. A, et al. Drosophila RNAi screen identifies host genes important for influenza virus replication. Nature. 2008;454:890–893. - PMC - PubMed
    1. Sabin L. R, Zhou R, Gruber J. J, Lukinova N, Bambina S, et al. Ars2 regulates both miRNA- and siRNA-dependent silencing and suppresses RNA virus infection in Drosophila. Cell. 2009;138:340–351. - PMC - PubMed
    1. Kim D. H, Feinbaum R, Alloing G, Emerson F. E, Garsin D. A, et al. A conserved p38 MAP kinase pathway in Caenorhabditis elegans innate immunity. Science. 2002;297:623–626. - PubMed
    1. Powell J. R, Kim D. H, Ausubel F. M. The G protein-coupled receptor FSHR-1 is required for the Caenorhabditis elegans innate immune response. Proc Natl Acad Sci U S A. 2009;106:2782–2787. - PMC - PubMed
    1. Troemel E. R, Felix M. A, Whiteman N. K, Barriere A, Ausubel F. M. Microsporidia are natural intracellular parasites of the nematode Caenorhabditis elegans. PLoS Biol. 2008;6:2736–2752. doi: 10.1371/journal.pbio.0060309. - DOI - PMC - PubMed

Publication types

MeSH terms

LinkOut - more resources