(A) The lateral seam cells (V1-4 and V6) of wild type animals execute stage-specific division patterns. At the L1, L3 and L4 stages, each seam cell executes a single asymmetric division, producing an anterior cell that fuses with the hyp7 syncitium, and a posterior daughter seam cell. At the L2 stage, a different pattern occurs: each seam cell divides symmetrically, and each daughter seam cell immediately divides again asymmetrically (red bars). This L2-specific duplicative seam cell division essentially doubles the number of seam cells in the hypodermis of C. elegans. In animals triply mutant for mir-48, mir-241, and mir-84, a reiteration of L2-specific seam cell division patterns occurs at the L3 stage. In contrast, animals over-expressing of miR-48 skip most L2 seam cell division programs, precociously execute vulval divisions and the produce adult specific cuticular structures at the end of the L3 stage. (B) nhl-2 encodes a member of the TRIM-NHL family of proteins and the ok818 deletion is a null allele of nhl-2 (Fig. S1). (C) In wild-type animals, col-19::GFP expression is adult-specific, and occurs in lateral seam and in hyp7 cells. In doubly-mutant mir-48; mir-84 adults, col-19::GFP expression is reduced in hyp7, and limited primarily to seam cells (arrows). In nhl-2; mir-48; mir-84 triply-mutant adults, hyp-7 col-19::GFP expression is reduced further, and other heterochronic phenotypes occur, including a reiteration of L2-specific seam cell division patterns in some V-lineages. (D and E) nhl-2 mutations similarly enhance the col-19::GFP expression phenotypes of let-7 mutants. (F and G) The precocious vulval induction, seam cell fusion and adult alae formation phenotypes associated with miR-48 over-expression are suppressed by removing nhl-2. Asterisks indicate that lateral seam cells of veIs48 animals precociously exit the cell cycle after the L3 molt. (H) Northern analysis of miRNAs derived from L2 molt wild-type, veIs48, and nhl-2(0);veIs48 animals. U6 RNA indicates loading controls. (I) nhl-2(0) enhances the L2-specific seam cell phenotypes associated with ain-1(ku322) mutations and also enhances L4-to-adult transition defects in alg-1(tm369) animals. Arrows indicate seam cell nuclei, solid lines indicate areas of cuticle containing adult-specific alae, and dashed lines indicate areas of cuticle lacking alae. (J) RNAi depletion of ain-1 in nhl-2(0) animals prevents the posttranscriptional down regulation of hbl-1::GFP. Arrows indicate the nuclear expression of the hbl-1::GFP reporter in hyp7 cells of L3 nhl-2(0); ain-1(RNAi) animals. (K) Proposed lineage diagrams that illustrate the synthetic phenotypes associated with nhl-2(0) and mutations of let-7-family miRNAs or of core miRISC components. Red bars indicate L2 specific cell division programs and green bars indicate those associated with L3 molt-L4 seam divisions.

(A) A diagram of C. elegans larva illustrating the location of ASEL and ASER neurons. lsy-6 mutants result in an absence of lim-6^pro::GFP expression in ASEL neurons. (B) Quantification of lim-6^pro::GFP mis-expression phenotypes in lsy-6, nhl-2 and nhl-2; lsy-6 compound mutants. (C) Ectopic expression of the lsy-6 miRNA from the cog-1 promoter results in the inappropriate down-regulation of COG-1::GFP in vulval and uterine tissues. Removing nhl-2 in the context of ectopic lsy-6 expression restores COG-1::GFP expression in uterus (Vul=vulva and u=uterus). (D) Quantification of uterine COG-1::GFP expression in various mutants. Brackets in B and D indicate statistically significant differences calculated from a two-tailed chi-squared analysis.

(A) cgh-1(ok492); mir-48(n4097);mir-84(n4037) lack adult alae (yellow line in wild type panel (left)) and contain supernumerary seam cells (arrowheads). (B) cgh-1(ok492); mir-48(n4097);mir-84(n4037) animals display a highly penetrant, vulval bursting phenotype that is not displayed in cgh-1(ok492) or mir-48(n4097); mir-84(n4037) mutants. (C) Animals were hatched onto hbl-1 RNAi food and assayed for adult specific col-19::GFP expression in early adulthood. (D) RNAi of cgh-1 in nhl-2(0) mutant backgrounds results a reiteration of L2-specific seam cell division patterns and adult nhl-2(0); cgh-1 RNAi animals display supernumerary seam cells and lack alae. (E and F) nhl-2(0); cgh-1 RNAi animals display defects in the expression of the adult-specific col-19::GFP transcriptional reporter and burst at the L4 molt. Mutations of hbl-1 dramatically reduce the penetrance of heterochronic phenotypes associated with cgh-1 RNAi in nhl-2(0) animals including vulval bursting. (G) The hbl-1 3′UTR GFP reporter is mis-expressed in nhl-2(0); cgh-1(RNAi) animals. (H) cgh-1(0) nhl-2(0) mutants also display a synthetic lethal phenotype during embryogenesis and larval development. (I) Lineage diagrams illustrating defects in temporal seam cell division programs of let-7-family miRNA mutants, compound cgh-1(0); let-7-family miRNA mutants, and nhl-2(0) cgh-1(0) mutants. Red bars indicate L2 specific cell division programs.

(A) The major components of the vulval induction pathway (See text for details). Components that are genetically involved in vulval induction are colored green where as those involved in Notch/lateral inhibition are colored blue. The let-7-family miRNA miR-84 (Red) is temporally expressed in presumptive 2° vulval cells. (B–E) nhl-2(0) mutations enhance let-60(n1046gf), Muv phenotypes. (F) nhl-2 mutations enhance let-60(ga89) temperature sensitive Muv phenotypes. Over expression of miR-84 suppresses the Muv phenotypes of let-60(ga89). The suppression of let-60(ga89) Muv phenotypes by miR-84 is reduced in animals lacking nhl-2. Brackets indicate statistically significant differences in phenotype calculated from a two-tailed chi-squared analysis from data derived from Table S3.

(A) A GFP::NHL-2 transgene is expressed in a variety of embryonic and larval tissues throughout development. (B) The cgh-1 genomic locus, with the location of the ok492 mutation indicated, and CGH-1 protein domain configuration, with ORF sequences identified in the NHL-2 2-hybrid screen (dashed lines). (C) Growth phenotypes of yeast strains co-transformed with various bait and prey constructs on non-selective or selective medium. (D and E) CGH-1::GFP and dsRED::NHL-2 co-localize in somatic tissues to several punctate, cytoplasmic bodies. dsRED::NHL-2 also co-localizes a translational DCAP-1::GFP transgene encoding the resident somatic P-body component.

(A) miRNA Northern analysis of total RNA extracted from wild-type, nhl-2(0), negative control RNAi, cgh-1 RNAi or an alg-1(0) mutant. Only mutations of alg-1, encoding the miRNA-associated Argonaute and core miRISC component, lead to defects in miRNA biogenesis (miss-processing and under-accumulation of mature miRNAs). (B) Immunoprecipitation of GFP::NHL-2 from extracts larvae co-precipitates ALG-1 and ALG-2. (C) Immunoprecipitation of CGH-1 co-precipitates ALG-1/2, AIN-1, and NHL-2. RNAse treatment of immunoprecipitated material reduces CGH-1 binding to core miRISC components but does not dramatically effect its interaction with NHL-2. (D) Co-immunoprecipitation of ALG-1/2 and AIN-1 by CGH-1 antiserum is not affected in the absence of nhl-2. (E) A model representing potential roles for NHL-2 in the post-transcriptional regulation of miRNA targets. We suggest that miRISC has a basal level of activity in the absence of NHL-2 and can lead to the reduction of protein products derived from miRNA target transcripts. Our models suggest that NHL-2 augments this basal level of activity by physically associating with miRISC. Proposed points of function and potential activities NHL-2 include 1) an enhancement of CGH-1-miRISC-target interaction by increasing the stability or fidelity of miRNA:target interactions; 2) a stimulatory role for NHL-2 in modulating intrinsic activity of CGH-1 and/or miRISC components; and the 3) a role for NHL-2 independent of CGH-1 or other known miRISC components.