© 2005 Nature Publishing Group
RNA interference is an antiviral defence mechanism
in Caenorhabditis elegans
Courtney Wilkins1, Ryan Dishongh1, Steve C. Moore1,3, Michael A. Whitt4, Marie Chow1& Khaled Machaca2
defined genetically in Caenorhabditis elegans1–4. RNAi has been
postulated to function as an adaptive antiviral immune mechan-
ism in the worm, but there is no experimental evidence for this.
Part of the limitation is that there are no known natural viral
pathogens of C. elegans. Here we describe an infection model in
C. elegans using the mammalian pathogen vesicular stomatitis
virus (VSV) to study the role of RNAi in antiviral immunity. VSV
mutants (rde-1; rde-4), leading to the production of infectious
response (rrf-3; eri-1). Because the RNAi response occurs in the
absence of exogenously added VSV small interfering RNAs, these
results show that RNAi is activated during VSVinfection and that
RNAi is a genuine antiviral immune defence mechanism in the
Vesicular stomatitis virus (VSV), an enveloped virus with a non-
segmented, negative-sense RNAgenome, was an attractive candidate
toinfect C.elegans cells.VSVis particularly notable for itsbroad host
range, including both mammalian and insect hosts5, and for its
ability to infect and replicate at temperatures used in culturing
C. elegans (18–258C). In addition, as the prototypical member of
stranded RNA viruses, its life cycle has been extensively studied in
mammalian cell culture6and VSV recombinant viruses have been
used as vectors for gene expression, vaccines and cell targeting7–10.
A replication-competent, recombinant VSV (Fig. 1a) expressing
the enhanced gene encoding green fluorescent protein (VSV-GFP)
was able to infect primary cell cultures isolated from wild-type (N2)
worms (Fig. 1b, VSV). GFP expression requires infectious virus
because no fluorescent signals were detected in cells infected with
virus that had been irradiated with ultraviolet (Fig. 1b, UV-VSV). In
addition, antibodies against the VSV nucleocapsid (N) and glyco-
protein (G) recognized only VSV-infected cells and not mock-
infected cells (Fig. 1c). By using primers specific to N gene coding
sequences in a reverse transcriptase-mediated polymerase chain
reaction (RT–PCR) assay specific for positive-sense viral RNAs, a
product was detected in VSV-infectedcells (Fig. 1d,lane2), but not in
uninfected cells orcells infected with UV-VSV(Fig. 1d, lanes 1 and 3).
The N gene RT–PCR signal, coupled with the presence of GFP and
viral proteins (Fig. 1b, c), confirmed that synthesis of the viral
transcripts was occurring within infected worm cells. Strand-specific
infected cells with the use of a primer set (P–M) amplifying the
intergenic region between the P and M genes from negative-sense
polymerase synthesizes a full-length, positive-sense RNA (anti-
genome) species that is complementary to the viral genome and
serves as a template for the synthesis of additional progeny RNA
antigenome and progeny genome RNAs and in a small fraction of
subgenomic, read-through messenger RNAs the P–M RT–PCR
product from infected samples indicates that increased levels of
viral genomes are present above that found in the initial inoculum
of virus. Thus viral replication as well as viral gene expression was
occurring in the infected C. elegans cells.
the worm. RNAi is triggered by processing double-stranded RNAs
Figure 1 | VSV infection of C. elegans cells in culture. a, Genome
organization of VSV-GFP recombinant virus. The structural proteins of the
VSV virion are the nucleocapsid (N), phosphoprotein (P), matrix (M),
after infection. Scale bar, 7mm. b, GFP fluorescence images (left) and
differential interference contrast images (right) of cells. c,
Immunofluorescent staining of cells for VSV-N or VSV-G proteins. d, RT–
PCR analyses for N gene sequences, the P to M intergenic region of the VSV
genome and C. elegans actin.
1Departments of Microbiology and Immunology and2Physiology and Biophysics, University of Arkansas for Medical Sciences, Little Rock, Arkansas 72205, USA.3Department of
Biology, Harding University, Searcy, Arkansas 72149, USA.4Department of Molecular Sciences, University of Tennessee Health Science Center, Memphis, Tennessee 38163,
Vol 436|18 August 2005|doi:10.1038/nature03957
© 2005 Nature Publishing Group
(dsRNAs) into small interfering RNAs (siRNAs). These siRNAs
the sequence-specific cleavage of target mRNAs and to induce post-
transcriptional gene silencing (PTGS)2–4. Genetic screens in the
worm have identified several genes involved in this process11,12;
among them are rde-1 and rde-4 as potential modulators of viral
immunity. RDE-1 is a member of the argonaute gene family11and
RDE-4 is a dsRNA-binding protein that interacts with both the
trigger dsRNA and RDE-1 (ref. 13). The RDE-4–RDE-1 complex
functions at the initiation step of RNAi, where it is thought to
recognize dsRNA and target it for cleavage into siRNAs by Dicer13.
Null mutants in rde-1 or rde-4 are incapable of inducing RNAi but
4 mutant cellsresulted in a higher percentage of GFP-positive cells in
the culture (Fig. 2a, b). In addition, the mean GFP fluorescence
intensities within individual cells continued to increase over time in
infected rde cells (Fig. 2c). Because GFP is expressed from the fourth
cistron, it is one of the least abundantly expressed proteins in VSV-
underestimate of the total percentage of infected cells within the
culture. To increase the sensitivity and obtain a more accurate
assessment, infected cells were also immunofluorescently stained
for N protein (the most abundant viral protein in an infected cell)
and analysed by flow cytometry. At 36h after infection, the percen-
tage of infected rde-1 cells (61%) in the culture was again about
double that for N2 cells (33%). Consistent with the increased
numbers of infected cells, there were greater amounts of both N
and G proteins immunoprecipitated from radiolabelled infected rde
cell lysates than from N2 cell lysates (Fig. 2d).
To determine whether VSV infection is productive in C. elegans,
N2 and rde cells were infected with VSV and the production of
infectious virus was followed over time (Fig. 2e). A rapid and
significant increase in viral titres was seen in infected rde cells
between 24 and 30h after infection, which reached a plateau by
36h after infection (Fig. 2e). The kinetics of virus growth in the rde
mutants is consistent with that normally observed during a synchro-
nized single cycle of productive viral infection. In contrast, no
significant increase in viral titres was detected at late times of
infection in the culture supernatants of infected N2 cells. The small
but continuous increase in virus titres observed over the entire
infection period (0–48h after infection) in N2 cells indicated that
this titre might be due primarily to the elution of virus particles off
Figure 2 | Increased infection of RNAi-deficient mutants. a–c, GFP
expression in VSV-GFP-infected N2, rde-1 (ne219) and rde-4 (ne301) cells.
a, Confocal images at low magnification (top row) and high magnification
positive cells (b) and intensities (four to eight individual fields) of GFP
fluorescence per cell (c) at each time after infection. Black squares, N2; red
circles, rde-4; blue triangles, rde-1. Results are means ^ s.e.m.
d, Immunoprecipitation (IP) of VSV N and G proteins from equivalent
numbers of infected N2 and rde-4 cells. Con, control. e, Virus titres from
infected N2 cells (black squares) and rde-1 cells (red circles). Similar results
were obtained with infected rde-4 cells. Results are means ^ s.e.m.
NATURE|Vol 436|18 August 2005
© 2005 Nature Publishing Group
cells from the initial inoculum, and that no detectable amounts of
new virus were produced in these cells.
Therefore, in an RNAi-null background, VSV infects a higher
percentage of cells than inwild-type N2 cells, and both viral proteins
and the GFP reporter accumulate to higher levels. Most significantly,
VSV infections in rde mutants result in productive viral infections
with significant increases in viral titres, which are absent from wild-
type N2 cells. These data show that the RDE-1 and RDE-4 com-
ponents of the RNAi pathway are important in inhibiting VSV
infection, which limit the extent of viral gene expression and the
production of infectious virus.
If an RNAi-null background potentiates viral infection, then
mutations that enhance RNAi responses would be predicted to
attenuate viral infection. Recently two worm mutants (rrf-3 and
eri-1) have been identified inwhich RNAi responses are enhanced by
modulating intracellular siRNA levels14,15. RRF-3, a member of the
RNA-dependent RNA polymerase (RdRP) gene family in C. elegans,
seems to inhibit RdRP-directed siRNA amplification, and worms
with mutations in rrf-3 are more sensitive, especially in neurons, to
RNAi responses induced by dsRNAs14. ERI-1, a member of the
DEDDh nuclease family, preferentiallycleaves siRNAs. Thus, siRNAs
are more stable and accumulate in eri-1 mutants, resulting in
enhanced gene suppression15. Comparison of VSV infection in rrf-
N2 cultures (Fig. 3a). However, a small subpopulation of eri-1 and
rrf-3 cells (about 2%) remained permissive to VSV infection, and
cells. This indicates that the RNAi response might remain negatively
regulated in this subpopulation of rrf-3 and eri-1 cells, thus allowing
in the whole worm in eri-1 mutants, in which some neurons remain
refractory to the RNAi response15. This virus-permissive phenotype
could be explained either by the differential expression of the two
presence of other as yet unidentified host suppressors of RNAi.
Nevertheless, the decreased numbers of GFP-positive cells indicate
that VSV infection is attenuated in host backgrounds displaying an
enhanced RNAi response.
The phenotypes observed on infection of N2 and the RNAi-
enhanced mutant cells indicate that the RNAi pathway might be
induced on viral infection without the exogenous addition of
synthetic siRNAs. RNase protection assays were performed to deter-
mine whether siRNAs were indeed generated as a consequence of
viral infection (Fig. 4). Protected bands of about 20–30 nucleotides
wereobserved when RNA frominfected N2 cells was hybridized with
radiolabelled VSV probes. In contrast, no protectionwas observed in
reactions using RNA from mock-infected N2 or virus-infected rde-4
cells. The presence of virus-specific siRNAs in infected N2 cells
indicates that the RNAi response is induced on virus infection.
Collectively, the genetic and biochemical data in both RNAi-null
and enhancedhost backgrounds establish RNAi as a cellular antiviral
defence mechanism in C. elegans. Pretreatment of mammalian cells
with synthetic siRNAs against viral sequences, to initiate the RNAi
pathway artificially, limits infection by several viruses including
polio, influenza and HIV16–18. This, coupled with the recognized
role of RNAi in antiviral immunity in plants19–21and insects22,23,
indicates that RNAi might serve a similar function throughout
evolution. Moreover, we show here that, similarly to infections
as a consequence of VSV infection. These data, together with recent
evidence that HIV infection can induce RNAi responses in human
cells24, indicate that activation of the RNAi pathway might be a
routine physiological response of the host to viral infections. Indeed,
many viruses, both plant and animal pathogens, have evolved
suppressors of RNAi to bypass this cellular immunity and enable
The relative genetic simplicity of C. elegans and the availability of
an extensive collection of mutants makes the worm an attractive
model host for identifying genetic determinants that affect the
susceptibility or resistance of a host to viral infection. In addition
to viral determinants, susceptibility to infection and disease is
dependent on the genetic background and physiological status of
Figure 3 | Increased resistance of RNAi-hypersensitive mutants to VSV
infection. Cells from N2 (squares), eri-1(mg 336) (filled circles) and rrf-
3(pk1426) (triangles) strains were infected with VSV-GFP, and GFP
expression over time were analysed by confocal microscopy. a, Percentages
of GFP-positive cells. b, Average intensities of GFP fluorescence per GFP-
positivecell.Resultsaremeans ^ s.e.m.fortwotofourindependentfieldsat
each time after infection.
Figure 4 | Virus-specific siRNAs in infected N2 cells. RNAs from mock-
infected N2, VSV-infected N2 and VSV-infected rde-4 cells at 24h after
infection were analysed by RNase protection assays with the use of a
radiolabelled probe specific for VSV N.
NATURE|Vol 436|18 August 2005
© 2005 Nature Publishing Group Download full-text
new model system for identifying genes involved in regulating
resistance or susceptibility not only to VSV but also to a wide
range of viruses, using VSV-based vectors and pseudotyped viruses.
The RNAi pathway is an example of such a set of host susceptibility
Primary cell culture of C. elegans. Cells were isolated as described in ref. 27,
with slight modifications. Early embryos were isolated after bleach treatment of
cultures of synchronized adult hermaphrodites on sucrose step gradients. The
embryos were incubated with chitinase (5Uml21) and chymotrypsin
(10mgml21). Monolayer cultures of the resultant dissociated cells were grown
at 258C in a modified L15 medium containing 10% fetal bovine serum, 30mM
sucrose and 100Uml21penicillin/streptomycin.
VSV infection. Virus stocks were grown in baby hamster kidney (BHK-21) cells
at low multiplicity (multiplicity of infection (m.o.i.) 0.01 plaque-forming units
per cell) and infectious titres were measured by plaque assay on HeLa cells at
378C. C. elegans cultures were infectedat high multiplicity (m.o.i. 10) byadding
with fresh medium. For virus growth curves, aliquots of the culture medium
were taken at intervals of 6h and viral titres were determined by plaque assayon
Immunoprecipitation. Cell monolayers (4 £ 106cells) were labelled from 42 to
48h after infection with [35S]methionine (100mCiml21), then harvested and
lysed in RIPA buffer. Samples were incubated with monoclonal antibodies
recognizing VSV N or G proteins, and Staph-G-conjugated beads were used to
collect the antibody complexes. Samples were analysed by 10% SDS–PAGE and
RT–PCR analyses. RNA was isolated from infected cells with the use of Trizol.
with specific RT primers that hybridized to viral sequences of either positive
the forward primer is complementary to nucleotides 730–747 and the reverse
primer is identical to nucleotides 1331–1365 of the minus-sense viral genome.
For the P–M intergenic region, the forward primer is complementary to
nucleotides 1499–1526 and the reverse primer was identical to nucleotides
2548–2570 of the viral genome. For the actin gene of C. elegans, the primer
sequences are identical to nucleotides 124–146 (forward primer) and comp-
lementary to nucleotides 820–849 (reverse primer) of the coding sequence.
RNase protection. Radiolabelled probes were generated by T7 transcription of
cDNA plasmids of VSV N, P, M, G and L genes28. RNA fractions enriched for
small RNAs (less than 200 bases) were isolated from cells by using the mirVana
miRNA isolation kit (Ambion) and RNase protection assays used the mirVana
miRNA detection kit (Ambion) in accordance with the manufacturer’s proto-
cols. Samples were analysed on a 15% denaturing polyacrylamide gel.
Confocal analysis. Cell monolayers were either examined for GFP fluorescence
or processed for immunofluorescent staining by fixing them in 1% paraformal-
dehyde, permeabilizing them with 0.1% saponin plus 1% BSA and incubating
them with primary antibodies against VSV-N or VSV-G proteins and phyco-
erythrin-conjugated secondary antibodies at appropriate dilutions.
Image analyses were performed with MetaMorph software. GFP images were
thresholded and the percentage of GFP-positive cells and the intensity of GFP
fluorescence in individual cells were measured.
Received 25 January; accepted 27 June 2005.
1.Grishok, A. & Mello, C. C. RNAi (nematodes: Caenorhabditis elegans). Adv.
Genet. 46, 339– -360 (2002).
Hannon, G. J. RNA interference. Nature 418, 244– -251 (2002).
Tijsterman, M. & Plasterk, R. H. Dicers at RISC; the mechanism of RNAi. Cell
117, 1– -3 (2004).
Denli, A. M. & Hannon, G. J. RNAi: an ever-growing puzzle. Trends Biochem. Sci.
28, 196– -201 (2003).
Letchworth, G. J., Rodriguez, L. L. & Del cbarrera, J. Vesicular stomatitis. Vet. J.
157, 239– -260 (1999).
6.Rose, J. K. & Whitt, M. A. in Fields Virology (eds Knipe, D. M. & Howley, P. M.)
1221– -1244 (Lippincott Williams & Wilkins, Philadelphia, 2004).
Schnell, M. J., Buonocore, L., Kretzschmar, E., Johnson, E. & Rose, J. K. Foreign
glycoproteins expressed from recombinant vesicular stomatitis viruses are
incorporated efficiently into virus particles. Proc. Natl Acad. Sci. USA 93,
11359– -11365 (1996).
Roberts, A. et al. Vaccination with a recombinant vesicular stomatitis virus
expressing an influenza virus hemagglutinin provides complete protection from
influenza virus challenge. J. Virol. 72, 4704– -4711 (1998).
Boritz, E., Gerlach, J., Johnson, J. E. & Rose, J. K. Replication-competent
rhabdoviruses with human immunodeficiency virus type 1 coats and green
fluorescent protein: entry by a pH-independent pathway. J. Virol. 73,
6937– -6945 (1999).
10. Takada, A. et al. A system for functional analysis of Ebola virus glycoprotein.
Proc. Natl Acad. Sci. USA 94, 14764– -14769 (1997).
11.Tabara, H. et al. The rde-1 gene, RNA interference, and transposon silencing in
C. elegans. Cell 99, 123– -132 (1999).
12. Tijsterman, M., May, R. C., Simmer, F., Okihara, K. L. & Plasterk, R. H. Genes
required for systemic RNA interference in Caenorhabditis elegans. Curr. Biol. 14,
111– -116 (2004).
13. Tabara, H., Yigit, E., Siomi, H. & Mello, C. C. The dsRNA binding protein RDE-4
interacts with RDE-1, DCR-1, and a DExH-box helicase to direct RNAi in C.
elegans. Cell 109, 861– -871 (2002).
14. Simmer, F. et al. Loss of the putative RNA-directed RNA polymerase RRF-3
makes C. elegans hypersensitive to RNAi. Curr. Biol. 12, 1317– -1319 (2002).
15. Kennedy, S., Wang, D. & Ruvkun, G. A conserved siRNA-degrading RNase
negatively regulates RNA interference in C. elegans. Nature 427, 645– -649
16. Gitlin, L., Karelsky, S. & Andino, R. Short interfering RNA confers intracellular
antiviral immunity in human cells. Nature 418, 430– -434 (2002).
17. Ge, Q. et al. RNA interference of influenza virus production by directly
targeting mRNA for degradation and indirectly inhibiting all viral RNA
transcription. Proc. Natl Acad. Sci. USA 100, 2718– -2723 (2003).
18. Hu, W. Y., Myers, C. P., Kilzer, J. M., Pfaff, S. L. & Bushman, F. D. Inhibition of
retroviral pathogenesis by RNA interference. Curr. Biol. 12, 1301– -1311 (2002).
19. Li, W. X. & Ding, S. W. Viral suppressors of RNA silencing. Curr. Opin.
Biotechnol. 12, 150– -154 (2001).
20. Vance, V. & Vaucheret, H. RNA silencing in plants—defense and
counterdefense. Science 292, 2277– -2280 (2001).
21. Voinnet, O. RNA silencing as a plant immune system against viruses. Trends
Genet. 17, 449– -459 (2001).
22. Li, H., Li, W. X. & Ding, S. W. Induction and suppression of RNA silencing by an
animal virus. Science 296, 1319– -1321 (2002).
23. Adelman, Z. N. et al. RNA silencing of dengue virus type 2 replication in
transformed C6/36 mosquito cells transcribing an inverted-repeat RNA
derived from the virus genome. J. Virol. 76, 12925– -12933 (2002).
24. Bennasser, Y., Le, S. Y., Benkirane, M. & Jeang, K. T. Evidence that HIV-1
encodes an siRNA and a suppressor of RNA silencing. Immunity 22, 607– -619
25. Roth, B. M., Pruss, G. J. & Vance, V. B. Plant viral suppressors of RNA silencing.
Virus Res. 102, 97– -108 (2004).
26. Li, W. X. et al. Interferon antagonist proteins of influenza and vaccinia viruses
are suppressors of RNA silencing. Proc. Natl Acad. Sci. USA 101, 1350– -1355
27. Christensen, M. et al. A primary culture system for functional analysis of C.
elegans neurons and muscle cells. Neuron 33, 503– -514 (2002).
28. Stillman, E. A., Rose, J. K. & Whitt, M. A. Replication and amplification of novel
vesicular stomatitis virus minigenomes encoding viral structural proteins.
J. Virol. 69, 2946– -2953 (1995).
Acknowledgements We thank the Caenorhabditis Genetics Center for most of
the strains used in this study; M. Kaufmann for suggestions; and K. Mitchell for
technical assistance in the initial phases of this study. This work was funded in
part by UAMS Foundation research funds (M.C.), from the BRIN Program of the
National Center for Research Resources (S.C.M. and M.C.), and startup funds
Author Information Reprints and permissions information is available at
npg.nature.com/reprintsandpermissions. The authors declare competing
financial interests: details accompany the paper on www.nature.com/nature.
Correspondence and requests for materials should be addressed to M.C.
(email@example.com) or K.M. (firstname.lastname@example.org).
NATURE|Vol 436|18 August 2005