C. elegans rrf-1 mutations maintain RNAi efficiency in the soma in addition to the germline

Abstract

Gene inactivation through RNA interference (RNAi) has proven to be a valuable tool for studying gene function in C. elegans. When combined with tissue-specific gene inactivation methods, RNAi has the potential to shed light on the function of a gene in distinct tissues. In this study we characterized C. elegans rrf-1 mutants to determine their ability to process RNAi in various tissues. These mutants have been widely used in RNAi studies to assess the function of genes specifically in the C. elegans germline. Upon closer analysis, we found that two rrf-1 mutants carrying different loss-of-function alleles were capable of processing RNAi targeting several somatically expressed genes. Specifically, we observed that the intestine was able to process RNAi triggers efficiently, whereas cells in the hypodermis showed partial susceptibility to RNAi in rrf-1 mutants. Other somatic tissues in rrf-1 mutants, such as the muscles and the somatic gonad, appeared resistant to RNAi. In addition to these observations, we found that the rrf-1(pk1417) mutation induced the expression of several transgenic arrays, including the FOXO transcription factor DAF-16. Unexpectedly, rrf-1(pk1417) mutants showed increased endogenous expression of the DAF-16 target gene sod-3; however, the lifespan and thermo-tolerance of rrf-1(pk1417) mutants were similar to those of wild-type animals. In sum, these data show that rrf-1 mutants display several phenotypes not previously appreciated, including broader tissue-specific RNAi-processing capabilities, and our results underscore the need for careful characterization of tissue-specific RNAi tools.

Figure 1. Intestinal elt-2 RNAi induces prominent phenotypes in rrf-1 mutants.

The indicated C. elegans strains were raised on bacteria expressing elt-2 dsRNA and imaged by DIC microscopy on day 3 of adulthood. The rde-1 strain and the rde-1 strain expressing RDE-1 in the hypodermis (lin-26p::rde-1) were unaffected by RNAi, whereas all other strains tested show growth defects, a clear appearance, and an abnormal intestine. This experiment was repeated four times with ∼100 worms per strain with similar results. WT: wild-type N2 (A – Hansen lab, B – Tuck lab), rrf-1(pk1417).4x: rrf-1(pk1417) outcrossed 4 times to WT(A).

Figure 2. GFP reporters in the intestine of rrf-1 mutants are repressed by gfp RNAi.

The indicated C. elegans strains were raised on bacteria expressing gfp dsRNA, and imaged by DIC and fluorescence microscopy (overlays shown here) on day 1 of adulthood. rrf-1 mutants carried the pk1417 allele. (A) The lgg-1p::gfp::lgg-1 reporter is expressed in the intestine, pharynx, somatic gonad, and hypodermis. GFP expression is dramatically reduced upon treatment with gfp RNAi in both wild-type and rrf-1 animals. The exposure time for the GFP channel under control conditions was 15 ms and for gfp RNAi 25 ms. In the gfp RNAi micrographs, note that the pharynx appears largely unaffected by RNAi in both wild-type and in rrf-1 animals. See Figure S5B, D for additional information on transgene intensity. (B) The daf-16p::gfp::daf-16 reporter is expressed in the intestine, and GFP expression is abolished in the wild-type and reduced in the rrf-1 background. The rrf-1 mutant shows visibly higher transgene expression levels compared to the wild-type strain. The exposure time for the GFP channel was 200 ms. See Figure S7A, B for additional information on transgene intensity. These experiments were repeated 2–3 times with 30–50 worms per strain and imaging of ∼10 per experiment, with similar results.

Figure 3. rrf-1 mutants are resistant to RNAi against the hypodermal gene bli-3, but are able to process gfp RNAi in seam cells.

The indicated C. elegans strains were raised on bacteria expressing bli-3 dsRNA (A) and gfp dsRNA (B) and imaged by bright-field or fluorescence microscopy, respectively, on day 1 of adulthood. (A) The rrf-1 mutants, as well as the RNAi-resistant rde-1 strain are unaffected by bli-3 RNAi, whereas all other strains tested show severe molting defects and varying degrees of blister formation. See Figure S4A for additional information. This experiment was repeated three times with 100–200 worms per strain with similar results, also for several generations. WT: wild-type N2 (A – Hansen lab, B – Tuck lab), rrf-1(pk1417).4x: rrf-1(pk1417) outcrossed 4 times to WT(A). (B) The SCMp::gfp reporter is exclusively expressed in hypodermal seam cells, which are denoted with arrows. GFP expression is abolished in the wild-type background and reduced in the rrf-1(pk1417) background. The rrf-1 mutant shows visibly higher GFP expression levels compared to the wild-type strain. The exposure time for the GFP channel was 100 ms. See Figure S4C for quantification of transgene intensity. These experiments were repeated 3 times with 30–50 worms per strain and imaging of ∼10 per experiment, with similar results.

Figure 4. rrf-1 mutants have increased sod-3 mRNA levels, a normal lifespan, and normal thermo-tolerance.

(A) The sod-3 mRNA levels of a mixed population of C. elegans wild-type (WT, N2(A)) and rrf-1(pk1417).4x (rrf-1(pk1417) mutant outcrossed four times to WT(A)) strains were determined. Bars show the mean + SEM of three independent experiments; **P<0.005 (Student’s t-test). (B) Lifespan analysis of wild-type WT(A) and rrf-1(pk1417).4x strains at 20°C; P = 0.59, log-rank (Mantel-Cox test). This experiment has been performed three times with similar results; see Figure S7F for additional data. (C) Thermo-tolerance was measured by assessing survival of WT(A) and rrf-1(pk1417).4x strains after 8 h incubation at 36°C; P = 0.36 (Student’s t-test). This experiment has been performed three times with similar results.