Schematic representation of NHL-2 NHL domain.

A) The overall NHL domain structure is shown in blue ribbon representation from the “top” of the 6-bladed beta propeller and from the “side” (PDBID:8SUC). Also shown are electrostatic surface potential diagrams from the same orientations. B) The dmBrat NHL (PDBID: 4ZLR) and C) drLIN41 (PDBID: 6FQ3) structures solved in the presence of RNA are shown alongside in comparable styles. D) The overall NHL domain structure is shown in blue ribbon highlighting the 6 blades and the position of 5 amino acids Lys817, Arg846, Lys910, Tyr935, Arg978. E-I) Overlay of NHL-2 and Brat NHL domains highlighting amino acids potentially impacting RNA binding. NHL-2 amino acids (blue sticks; PDBIB:8SUC) and dmBrat NHL amino acids (tan sticks) with bound RNA (green sticks) (PDBID: 4ZLR).

X-ray data statistics

Tyr935 and Arg978 are essential for RNA-binding.

Representative fluorescence anisotropy binding curves for wild-type NHL-2 NHL domain and mutant version to 17-mer U-rich RNA. Mut[1-5] (Lys817Ala, Arg846Ala, Lys910Ala, Tyr935Ala, and Arg978Ala), Mut[1-3] (Lys817Ala, Arg846Ala, Lys910Ala) and Mut[4-5] (Tyr935Ala, and Arg978Ala). Data points represent triplicate samples with standard error of the mean (SEM). The binding data were fit by the 1:1 Langmuir binding model to derive the KD.

Analysis of genetic interactions of nhl-2(ΔRING) and nhl-2(RBlf) mutants.

A) Quantification of alae defects in nhl-2(ΔRING) and nhl-2(RBlf) worms with and without cgh-1(RNAi) B) Representative DIC images of normal and defective alae. C) Analysis of col-19::GFP reporter expression in wild-type, nhl-2(RBlf) and nhl-2(ok818) worms. D) Quantification of col-19::GFP reporter expressing cells counted in WT worms and nhl-2 (ok818), nhl-2(RBlf) and effect of cgh-1(RNAi) on the strains.

Synthetic interaction of nhl-2 mutants with selected miRISC cofactors.

A) Schematic diagram of experimental design. B) Quantification of alae formation in wild-type, nhl-2(ΔRING) and nhl-2(RBlf), nhl-2(ok818) worms were determined when specific miRISC cofactors were knocked down. Alae were scored normal or defective (combined mis-shaped and no alae). Error bars represent the standard deviation.

Putative NHL-2 binding sites are required for lin-28 3’ UTR Reporter regulation.

A) Schematic representation of the PEST-mCherry-H2B reporter constructs. The pcol-10 promoter drives expression of the PEST::mCherry-H2B reporter in seam cells and translation regulated via the lin-28 3’ UTR. The let-7 binding site is shown in brown. The wild-type lin-28 3’ UTR contains three putative NHL-2 binding sites which were mutagenized by addition of the cytosine into the middle of the binding site. B) Analysis of lin-28 3’ UTR mCherry reporter constructs in seam cells of wild-type and nhl-2(ok818) worms.

P bodies are larger in NHL-2(RBlf) worms.

A) GFP::NHL-2 and GFP::NHL-2(RBlf) co-localise with mCherry::DCAP-1 in P-bodies, however GFP::NHL-2(RBlf) P-bodies are larger compared to GFP:: NHL-2. B) Quantification of P-bodies in GFP::NHL-2 and GFP::NHL-2(RBlf). P-body sizes were measured in 15 adult worms in the tail area. P value <0.0001. C) Quantification of P-body size in L4 worms expressing GFP::NHL-2 and NHL-2(= RBlf) subjected to RNAi knockdown of specific RNA decay factors. Images were analysed using Fiji software to determine P-body size. Data were analysed using one-way ANOVA multiple by GraphPad Prism 9.5. P-bodies were measured in 5 worms in each strain.

let-7 target mRNAs are enriched in ALG1/2 IPs.

Fold of enrichment of let-7 target mRNAs daf-12, hbl-1, let-60 and lin-41 in the ALG1/2 IPs and total RNA from L4 stage wild-type (WT), nhl-2(ok818), cgh-1(RNAi) and nhl-2(0ok818);cgh-1(RNAi) worms. For total RNA samples, mRNA levels were quantified by qRT-PCR and normalized to act-1 mRNA. For RNA isolated from the ALG-1/ALG-2 IPs, fold enrichment of a given mRNA was determined by comparing the normalized concentration of this mRNA in the IP sample to the corresponding input worm lysate sample. Error bars represent the 95% confidence intervals calculated from samples using the ΔΔCt method.

IFET-1 and CGH-1 are predicted to physically interact.

A) Alignment of the Cup Homology Domain (CHD) from C. elegans (Ce), D. melanogaster (Dm), D. rerio (Dr), X. laevis (Xl), M. (Mm) and H. sapiens (Hs). Conserved residues are highlighted in red and human 4E-T amino acids that interact with DDX6C are indicated with red asterisks. B) AlphaFold2 structure prediction of the IFET-1-CGH-1 interaction. The CGH-1 Rec2A (amino acids 246-420) is shown in aqua, and the CHD of IFET-1 is shown in green with predicted CGH-1 interacting amino acids highlighted in red.

A model of function of NHL-2/CGH-1/IFET-1 effector complex in miRNA-mediated silencing.

A) In wild-type worms NHL-2 binds to selected miRISC target mRNAs and directly interacts with CGH-1, which in turn directly binds to IFET-1, resulting in inhibition of translation initiation. The direct interaction between IFET-1 and IFE-3 provides a mechanism for selecting subsets of miRNA targets 5′ cap structure. Conserved protein-protein interactions with the NHL-2/CGH-1/IFET-1 effector complex results in the recruitment of the deadenylation and decapping complexes which then mRNA decay. B) In the absence of two of the three members of the NHL-2/CGH-1/IFET-1 translational inhibition effector complex the miRNA target remains available for translation.

Analysis of genetic interactions of nhl-2(ΔRING) and nhl-2(RB MUT1-5) mutants.

A) Illustration of the mutations made to the RING and NHL domains. The RING domain was rendered non-functional RING domain by replacing Cys77, Cys82, and Cys85 with alanine (red circles). RNA-binding activity was inhibited by replacing K817, R846, K910, Y935, and R978 with alanine). B) Quantification of alae defects in nhl-2(RING) and nhl-2(RB MUT1-5) worms with and without cgh-1(RNAi).

Volcano plot protein enrichment from mass spectrometry analysis of ALG-1/ALG-2 immunoprecipitations.

IFET-1 is significantly enriched in wild-type compared to nhl-2(ok818) worms.