A role for the RNase III enzyme DCR-1 in RNA interference and germ line development in Caenorhabditis elegans

Abstract

An early event in RNA interference (RNAi) is the cleavage of the initiating double-stranded RNA (dsRNA) to short pieces, 21 to 23 nucleotides in length. Here we describe a null mutation in dicer-1 (dcr-1), a gene proposed to encode the enzyme that generates these short RNAs. We find that dcr-1(-/-) animals have defects in RNAi under some, but not all, conditions. Mutant animals have germ line defects that lead to sterility, suggesting that cleavage of dsRNA to short pieces is a requisite event in normal development.

Fig. 1

dcr-1 mutants are deficient for RNAi. (A) The gene structure of dcr-1 (K12H4.8, chromosome III) is shown with a dotted line indicating the deletion found in dcr-1(ok247). The deletion extends from nucleotides 3522 to 5991, relative to the ATG codon. Regions encoding the RNA helicase domain (gray), the RNase III domains (blue), and the dsRNA binding motif (red) are underlined. To ensure that the observed phenotypes in dcr-1(ok247) animals were due to a mutation in dcr-1, we outcrossed animals heterozygous at the locus against wild-type animals (N2) eight times. (B) Wild-type ( WT ) or dcr-1(ok247) animals expressing a GFP transgene were compared by fluorescence imaging before (−RNAi) or after (+RNAi) heat shock to induce the GFP RNA hairpin (29). The sur-5::GFP transgene is expressed in many cells, with brightest expression in the intestine (30). GFP fluorescence in wild-type animals is reduced after the heat shock treatment, whereas dcr-1(ok247) mutants continue to exhibit high levels of GFP fluorescence comparable to non–heat-shocked animals. Even in wild-type animals, GFP fluorescence remains after heat shock in neuronal tissues, which are often refractory to RNAi. Heterozygous individuals behaved like wild-type animals, exhibiting a substantial loss in GFP fluorescence (24). Transgenic lines were constructed with standard protocols (31), and C. elegans strains were maintained as described (32).

Fig. 2

Analyses of mRNA levels show that dcr-1(−/−) mutants are defective for RNAi of some but not all genes. dsRNA corresponding to mpk-1 (A) or unc-54 (B) was injected into L4 animals (33), and endogenous mRNA levels were quantified for single worms by semiquantitative RT-PCR (34). Bars show average mRNA levels (n = 3) for wild-type (solid) and dcr-1(−/−) animals (hatched) with (+) or without (−) dsRNA injection. Error bars show SEM.

Fig. 3

Germ line development is abnormal in dcr-1(−/−) animals. (A through C) DIC micrographs. Distal (d) and proximal (p) orientation is indicated. (A) Wild-type adult hermaphrodite. The arrow points to a normal oocyte in the proximal gonad. (B) dcr-1 adult hermaphrodite with a burst vulva. An abnormal misshapen oocyte is indicated by an arrow in the proximal gonad (arrow). (C) dcr-1 adult hermaphrodite with enlarged oocyte nuclei (arrows). (D) Fluorescent micrograph of the DAPI-stained animal of (C). Areas of intense staining colocalize to nuclei (arrows). (E) Merged micrograph of images in (C) and (D).