C. elegans RIG-I-like receptor DRH-1 signals via CARDs to activate anti-viral immunity in intestinal cells

Upon sensing viral RNA, mammalian RIG-I-like receptors activate downstream signals using caspase activation and recruitment domains (CARDs), which ultimately promote transcriptional immune responses that have been well-studied. In contrast, the downstream signaling mechanisms for invertebrate RIG-I-like receptors are much less clear. For example, the Caenorhabditis elegans RIG-I-like receptor DRH-1 lacks annotated CARDs and upregulates the distinct output of RNA interference (RNAi). Here we found that, similar to mammal RIG-I-like receptors, DRH-1 signals through two tandem caspase activation and recruitment domains (2CARD) to induce a transcriptional immune response. Expression of DRH-1(2CARD) alone in the intestine was sufficient to induce immune gene expression, increase viral resistance, and promote thermotolerance, a phenotype previously associated with immune activation. We also found that DRH-1 is required in the intestine to induce immune gene expression, and we demonstrate subcellular colocalization of DRH-1 puncta with double-stranded RNA inside the cytoplasm of intestinal cells upon viral infection. Altogether, our results reveal mechanistic and spatial insights into anti-viral signaling in C. elegans, highlighting unexpected parallels in RIG-I-like receptor signaling between C. elegans and mammals.


Figures S1 to S7 Tables S1 to S3 SI References
Other supporting materials for this manuscript include the following:

C. elegans Maintenance
C. elegans strains were maintained on Nematode Growth Media (NGM) agar plates containing streptomycin-resistant Escherichia coli OP50-1, unless otherwise specified.S1 Table lists all strains used in this study.
Figure 1C and 1D demonstrate that DRH-1(2CARD)::mScarlet transgenic lines jyEx302, jyEx304 and jyEx305 all induce the pals-5p::GFP reporter.However, specific lines were used for specific subsequent experiments for the following reasons.First, developmental analysis of pals-5p::GFP expression (Fig. 1E) was performed with transgenic line jyEx302, as this was the first line DRH-1(2CARD) line established.Next jyEx305 was chosen for qRT-PCR (Fig. 1F), because it has the highest rate of transmission of the extrachromosomal array, increasing the percentage of transgenic animals in a large mixed population used to obtain enough RNA for analysis.This strain also has the lowest expression of mScarlet among all three transgenic lines, and for this reason we chose jyEx305 for Fig. 2 and S2, to minimize red fluorescence bleed-through into the green channel, in order to better image pals-5p::GFP and ZIP-1::GFP.All three lines increased resistance to viral infection (Fig. 3A, B), but only transgenic line jyEx304 was used for subsequent N. parisii infection and thermotolerance assays, as this line showed significantly decreased viral load, but also exhibited less developmental delay compared to jyEx305.Localization of DRH-1(2CARD) was assessed in transgenic line jyEx304 (Fig S4A , B) as this line showed significantly decreased viral load (Fig. 3B) while still providing a sufficient number of infected transgenic animals for analysis.
To create the empty vector control, the drh-1(2CARD) sequence was deleted from pET770 using the Q5 Site-Directed Mutagenesis Kit (New England Biolabs).Primers LEB003 and LEB004 were used to amplify around the drh-1(2CARD) sequence in pET770.The linearized fragment was subsequently recircularized and transformed into chemically-comptent NEB DH5a cells (New England Biolabs) to generate the resulting plasmid pET786 [vha-6p::mScarlet::unc-54 3'utr].10 ng/μl of pET786 was injected into jyIs8; rde-1 animals.Of note, the negative control construct was initially injected at 100 ng/μl, which is the injection concentration used for pET770 (see above).
However, injections at this concentration resulted in extremely high levels of mScarlet expression, in which red fluorescence levels greatly exceeded that of DRH-1(2CARD)::mScarlet lines.
Additionally, bleed-through of red fluorescence into the green channel used for imaging was observed in these animals, impairing assessment of pals-5p::GFP expression.Therefore, a decreased injection concentration of 10 ng/μl, was necessary to achieve protein expression levels comparable to that of DRH-1(2CARD)::mScarlet, as well as mitigate any off-target effects due to extremely high concentrations of mScarlet.Of note, the mScarlet expression in the negative control line was still higher than in the DRH-1(2CARD) lines.Filler DNA (pBluescript) was added to the injection mix to reach a final DNA concentration of 100 ng/μl.
For intestinal expression of full-length drh-1, an intron-containing sequence for drh-1 was amplified from plasmid Tian233 using primers LEB005 and LEB006, which contain a 40 bp 5' to vha-6p and a 36 bp 3' homology to mScarlet.A vector backbone was generated by linearization of pET770, and PCR products were assembled using the NEBuilder HiFi DNA Assembly kit (New England Biolabs) to generate the final construct pET788. 100 ng/μl of pET788 was injected into jyIs8; rde-1 animals.
All plasmid sequences were validated by whole-plasmid sequencing.Constructs used in this study can be found in S2 Table .Primer sequences are listed in S3 Table.

Synchronization of C. elegans
To synchronize development, gravid adult animals were washed off plates with M9 media and transferred into a 15 ml conical tube.Worms were pelleted at 3,000 rpm for 30 sec and resuspended in 3 ml of M9 and 1 ml of bleaching solution (500 μl of 5.65-6% sodium hypochlorite solution and 500 μl of 2 M NaOH).After most adults had dissolved, released embryos were washed five times with 15ml of M9 and resuspended in a final volume of 3 ml of M9.Embryos were incubated at 20 ºC with continual rotation for 20-24 h to hatch L1s.

pals-5p::GFP and F26F2.1::GFP fluorescence measurements
DRH-1(2CARD) strains were synchronized by bleaching and grown at 20 ºC for 24 h (L2/L3 larval stage), 44 h (L4 larval stage), or 68 h (adults).Worms were washed off plates, resuspended in M9, and anesthetized with 10 µM levamisole prior to imaging in a 96-well plate format on a ImageXpress Nano plate reader using the 4x objective (Molecular Devices, LLC, San Jose, CA).Backgroundcorrected mean fluorescence intensity was quantified in Fiji by tracing the worm area and calculating the mean fluorescence intensity (MFI).To correct for background fluorescence, MFI of the well was subtracted from the MFI of each worm.For tissue-specific RNAi experiments, GFP fluorescence was measured using a COPAS Biosort instrument (Union Biometrica).

RNA isolation and qRT-PCR
For DRH-1(2CARD) lines in Fig. 1F, 2C, S2C, and S4C, 40 adults were transferred to 10-cm NGM plates due to smaller brood sizes.For non-transgenic controls, 20 adults were transferred to 10cm NGM plates.As mentioned above, DRH-1(2CARD) lines were generated in a RNAi-defective rde-1 mutant background to enhance transgene expression.Thus, non-transgenic control strains are either rde-1 mutants with (Fig. S1B) or without (Fig. 1F) the pals-5p::GFP reporter.Two replicates (two plates) were set up per strain and worms were incubated for 96 h at 20 ºC to produce progeny.RNA was isolated from these progeny (thousands of animals) using TRI Reagent and 1bromo-3-chloropropane (BCP) (Molecular Research Center) followed by isopropanol and ethanol washes.Pure RNA was resuspended in nuclease free water.Following cDNA synthesis using the iScript synthesis kit (Bio-Rad), qPCR was performed with iQ SYBR Green Supermix (Bio-Rad) on a CFX Connect Real-Time PCR Detection System (Bio-Rad).In Fig. S4B, RNA was isolated 24 h after infecting synchronized rde-1 L4s with either Orsay virus or N. parisii at 25ºC.For measurements of drh-1 transcript levels in Fig. S5, synchronized jyIs8 L1s were plated on control or drh-1 RNAi plates and grown at 20ºC for 44 h following the protocol detailed below.L4 animals were exposed to virus for 24 h prior to RNA isolation and cDNA synthesis following the steps above.
For all qPCR analyses, relative gene expression ratios were determined by the Pfaffl method (7).
Primer efficiencies were calculated by generating standard curves.Gene expression values were normalized to the expression of housekeeping gene snb-1.qRT-PCR primer sequences are listed in S3 Table.

Bortezomib treatment
Proteasome inhibition by bortezomib (Selleck Chemicals) was performed as previously described (8).Synchronized L1 animals were plated on 6-cm NGM plates containing OP50-1 and grown for 44 h at 20 ºC.Bortezomib in DMSO was top-plated to reach a final concentration in the agar of 20 µM.DMSO was used as a vehicle control.Plates were then incubated for 4 h at 20 ºC prior to imaging.To assess ZIP-1::GFP localization, animals were anesthetized with 100 µM levamisole and imaged on a Zeiss LSM700 confocal microscope with Zen 2010 software.

Orsay virus infections
All infections were performed using virus from the same batch of virus filtrate, which was prepared as previously described (9).Animals were infected with virus filtrate at the L1 or L4 stage.A minimum of two replicates (two plates) were set up per genotype per experiment.At least three independent experiments were performed per timepoint.For L1 infections, developmentally synchronized animals were exposed to a mixture of OP50-1 bacteria, M9, and virus filtrate for 18 h at 20 ºC (Fig 3A, B).For L4 infections, synchronized animals were exposed to a mixture of OP50-1 bacteria, M9, and virus filtrate for 24 h at 20 ºC (Fig 4, 5).For infection rate measurements and localization analyses, animals were washed in M9 and fixed in 4% paraformaldehyde for 15 min.Following fixation, worms were incubated at 47 ºC overnight with fluorescein (FAM)-or Quasar 670-conjugated FISH probes that target Orsay virus RNA1 and RNA2 (Biosearch Technologies).
Infection rate was assessed using a Zeiss AxioImager M1 compound microscope.For each experimental replicate, a minimum of 100 animals per genotype were scored for the presence of FISH fluorescent signal.For subcellular localization analyses, animals were imaged using a Zeiss LSM700 confocal microscope with Zen 2010 software.For tissue-specific knockdown experiments, animals were anesthetized with 10 µM levamisole prior to analysis using a COPAS Biosort instrument (Union Biometrica) to measure fluorescence.For qRT-PCR analysis, see RNA isolation and qRT-PCR section.

Microsporidia infections
N. parisii spores were prepared as previously described (10).A mixed population of DRH-1(2CARD) animals and non-transgenic siblings was synchronized at the L1 stage and plated on 6cm NGM plates along with a mixture of OP50-1, M9, and one million spores for 30 h at 25ºC.A minimum of four replicates (four plates) were set up per treatment.Three independent experiments were performed.To assess the presence of N. parisii meronts, animals were washed in M9 and fixed in 4% paraformaldehyde for 15 min prior to an overnight incubation at 47 ºC with a FAMconjugated FISH probe that hybridizes to N. parisii ribosomal RNA (Biosearch Technologies).
Samples were sorted based on red fluorescence using the COPAS Biosort instrument (Union Biometrica) to obtain homogenous populations of either DRH-1(2CARD) animals or non-transgenic siblings.Green fluorescence was measured for each population using the COPAS Biosort instrument (Union Biometrica).For each genotype, the median green fluorescence of the infected population was normalized to the median fluorescence of the uninfected population.For qRT-PCR analysis, see RNA isolation and qRT-PCR section.

Development rate measurements
60 gravid adults were transferred onto a 10-cm NGM plate containing OP50-1 and incubated at 20 ºC for 2 h.Adults were gently washed off of plates with M9 such that only eggs remained on the plate, and eggs were incubated at 20 ºC for 48 h and 64 h.For each timepoint, development rate was determined by scoring the percentage of L4 stage or older animals.100 animals were scored per genotype across three independent experimental replicates.
Animals were exposed to heat-shocked for 2 h at 37.5 ºC in a dry incubator.Immediately following heat shock, plates were transferred to room temperature for 30 min.followed by incubation at 20 ºC for 24 h.Survival was scored by assessing mobility, wherein dead worms were identified by failure to respond to a single touch using a worm pick.Three independent experimental replicates were performed on different days.

Tissue-specific RNA interference
Systemic, intestine-specific, and epidermis-specific RNAi was performed via the feeding method.
RNAi bacterial clones (drh-1 and L4440 empty vector control) were inoculated in 5 ml LB containing 50 µg/ml carbenicillin and incubated in a 37 ºC shaking incubator (250 rpm) for 16 h.Overnight cultures were seeded onto NGM plates supplemented with 5 mM IPTG and 1 mM carbenicillin.
Seeded RNAi plates were incubated at room temperature for 4 days.Synchronized L1 animals were transferred to RNAi plates and grown at 20 ºC for 44 h.Animals were then exposed to virus for 24 h and then analyzed for pals-5p::GFP expression on a COPAS Biosort instrument (Union Biometrica), as described above.
Puncta/aggregate density measurements were calculated by dividing the number of puncta/aggregates by the total area of the annotated region for each animal.

Immunohistochemistry
For dsRNA localization analyses, synchronized L1s were plated onto 10-cm NGM plates and grown for 44 h at 20 ºC to reach the L4 stage.Animals were then exposed to virus filtrate for 24 h at 20 ºC, as described above.After 24 h, adults were washed off of plates with M9 and anesthetized in 100 µM levamisole.Intestines were dissected out of 100-200 adults and fixed in 4% paraformaldehyde for 15 min.To visualize virus-infected cells, samples were incubated with Quasar 670-conjugated FISH probes that hybridize to Orsay virus RNA1 and RNA2 (Biosearch Technologies).Following a 6-8 h incubation at 47 ºC, FISH probes were washed off and dissected intestines were incubated in block buffer (PBS, 0.5% Tween-20, 1 mM EDTA, 5% BSA, 0.05% NaN3) overnight at 4 ºC.Dissected intestines were then stained with a 1:200 dilution of the anti-dsRNA antibody clone rJ2 (Sigma-Aldrich) for 2-4 h, followed by 10 µg/ml of goat anti-mouse IgG(H+L) cross-adsorbed secondary antibody conjugated to Alexa Fluor 488 (Invitrogen) for 1-2 h.
Staining was performed in block buffer at room temperature.

Statistics
All statistical analyses were performed in R. Q-Q plots were used to assess normality of the data, and parametric tests were used when appropriate.A nonparametric test was applied to data that did not meet the assumptions for a parametric test.

Fig. S2 .
Fig. S2.DRH-1(2CARD) induces F26F2.1p::GFP and endogenous IPR gene expression.(A) Representative images showing increased F26F2.1p::GFPexpression in DRH-1(2CARD) transgenic animals relative to their non-transgenic siblings.Scale bar = 50 µm.(B) Quantification of F26F2.1p::GFP in DRH-1(2CARD) animals and non-transgenic siblings shown in (A).Horizontal lines in box-and-whisker plots represent median values, and the box reflects the 25 th to 75 th percentiles.Each dot represents an individual animal; 150 animals were analyzed across three independent experiments for each strain.A Mann-Whitney U test was used to determine significance; ****p < 0.0001.(C) qRT-PCR analysis of DRH-1(2CARD) line jyEx305 (expresses pals-5p::GFP reporter) shown in.RNA was extracted from a mixed-stage population containing both DRH-1(2CARD) transgenic animals and their non-transgenic siblings.Fold change in gene expression was determined relative to a non-transgenic control strain (rde-1 mutant in a pals-5p::GFP background).Bars represent the mean across experimental replicates; error bars represent the standard deviation.Each dot represents a biological replicate (a plate with a minimum of 2000 animals); four independent experimental replicates were performed.A one-tailed t-test was used to calculate p-values; *p < 0.05.

Fig. S3 .
Fig. S3.DRH-1(2CARD) expression does not promote nuclear localization of ZIP-1::GFP.(A) Representative images showing ZIP-1::GFP expression in DRH-1(2CARD) transgenic animals and non-transgenic siblings.ZIP-1::GFP is not visible in the nuclei of DRH-1(2CARD) animals.Bortezomib treatment was used as a positive control for nuclear localization of ZIP-1::GFP.White arrowheads indicate ZIP-1::GFP expression in the nucleus.Yellow arrowheads indicate autofluorescence from intestinal gut granules.Scale bar = 25 µm.(B) ZIP-1::GFP is not present in intestinal nuclei of untreated animals, but is expressed in 100% of animals treated with bortezomib.Localization pattern of ZIP-1::GFP is the same for both DRH-1(2CARD) animals and nontransgenic siblings.For each genotype and treatment, 45 total animals were scored.Bars represent the mean across biological replicates; error bars represent the standard deviation.Each dot represents a biological replicate (a plate with 15 animals).

Fig. S5 .
Fig. S5.RNAi against drh-1 results in decreased drh-1 mRNA transcript levels.qRT-PCR analysis of drh-1 confirms significantly decreased mRNA expression following RNAi against drh-1.Bars represent the mean across experimental replicates; error bars represent standard deviation.Each dot represents an experimental replicate; five independent experimental replicates were performed.A one-tailed t-test was used to determine significance; **p < 0.01.

Fig. S6 .
Fig. S6.Expression of the rpl-28p::mScarlet::drh-1 transgene rescues pals-5 expression and Orsay RNA1 levels upon viral infection.(A) During viral infection, expression of the mScarlet::DRH-1 transgene (mgTi54) in a drh-1(jy110) deletion mutant is sufficient to rescue mRNA levels of (A) IPR genes F26F2.1 and pals-5, as well as (B) Orsay RNA1 to WT levels.qRT-PCR analysis of F26F2.1, pals-5, and Orsay RNA1 in WT animals (N2), a drh-1(jy110) deletion mutant, and animals that express the mgTi54 transgene in a drh-1(jy110) or WT background.Fold change in gene expression was determined relative to the uninfected WT (N2) control.Bars represent the mean; error bars represent the standard deviation.Each dot represents a biological replicate (a plate with 2000 animals); three independent experimental replicates were performed.A one-sample Wilcoxon signed rank test was used to compare the distribution of values against a hypothetical value of 1 in the uninfected group.A Mann-Whitney U test was used to calculate p-values for comparisons between samples in the infected group; *p < 0.05.

Table S1 .
List of strains used in this study.

Table S2 .
Constructs used in this study.

Table S3 .
Primers used in this study.