Ataxin-2, Twenty-four, and Dicer-2 are components of a noncanonical cytoplasmic polyadenylation complex

Fly strains and generation of depleted and null embryos

OregonR and w¹¹¹⁸ were used as wild-type flies. The following mutant or transgenic stocks were used: w; Dcr2^L811fsX/CyO, amos^Roi-1 (kindly provided by Dr. Martine Simonelig), w; UAS-Atx2-RNAi (Vienna Drosophila Resource Center, VDRC #34955), w; UAS-Tyf-RNAi (VDRC #21965) and w; alphaTub67C-GAL4-VP16; alphaTub67C-GAL4-VP16 (containing the driver in both chromosomes II and III, provided by Dr. Jerome Solon). Flies were maintained on standard food at 25°C.

Dicer-2 null embryos were obtained by crossing w; Dcr2^L811fsX homozygous females with w¹¹¹⁸ males. For the generation of Atx2-depleted embryos, homozygous virgin w; UAS-Atx2-RNAi females were crossed with homozygous alphaTub67C-GAL4-VP16 males, and resulting females were then crossed with w¹¹¹⁸ males. A similar strategy was followed to obtain Tyf depleted embryos using w; UAS-Tyf-RNAi females.

Embryo extracts

For Dicer-2 interactome and RIP-Seq analysis, large-scale collections of staged OregonR 90 min embryos were obtained and processed as described in Coll et al (2010). For PAT assays, the protocol was scaled down as follows, carrying depleted/null and wild-type embryos in parallel. Egg-laying trays were exchanged three times for embryo synchronization, and 90 min embryos collected thereafter. In the case of Dcr-2 null embryos, because of their scarcity, multiple 90-min collections were taken in a period of 6–8 h, kept on ice, and pooled. Embryos were washed twice in EW buffer (0.7% NaCl, 0.04% TritonX-100), dechorionated in bleach (1:3 in EW buffer) for 2–3 min, and washed several times with milliQ water. Dechorionated embryos were then washed once with cold DEI buffer (10 mM Hepes, pH 7.4, 5 mM DTT, and 1× complete protease inhibitor from Roche) followed by homogenization in 1–2 embryo volumes of DEI buffer on ice. The homogenate was centrifuged at 10,000g for 10 min at 4°C. The intermediate cytosolic fraction was carefully collected, avoiding the upper lipid layer and the pellet. Glycerol was added to 10%, and extracts aliquoted, snap frozen, and stored at −80°C.

Antibodies

Rabbit αDicer-2 polyclonal antibodies were raised in house against the first 151 aa of the protein and affinity-purified using HiTrap NHS-activated HP columns (GE Healthcare) coupled with His-Dicer-2(1–257)-MBP (Coll et al, 2018). αWispy antibodies were kindly provided by Mariana Wolfner (Cui et al, 2008). αTyf and αAtx2 antibodies were provided by Joonho Choe (Lim et al, 2011) and Mani Ramaswamy (McCann et al, 2011), respectively. αBic-C antibodies were provided by Paul Lasko (Chicoine et al, 2007). αUNR antibodies have been described previously (Abaza et al, 2006). Antibody dilutions for Western blots were as follows: αDicer-2 (1:1,000), αTyf (1:500), αAtx2 (1:500), αWispy (1:2,000), αUNR (1:1,000), αBic-C (1:1,000).

RNA immunoprecipitation followed by sequencing (RIP-Seq)

Protein-A Dynabeads (Invitrogen) containing affinity purified αDicer-2 antibody to their maximum binding capacity (2.4 μg antibody per 10 μl beads) were prepared by washing beads with 10 volumes of IPP 500 buffer (20 mM Hepes, 500 mM NaCl, 1.5 mM MgCl₂, 0.5 mM DTT, and 0.05% NP-40) followed by incubation with αDicer-2 antibody for 30 min at room temperature. Antibody-bound beads were washed twice with 10 vol of IPP 500 and IPP containing 150 mM NaCl (IPP 150). A similar procedure was followed to obtain beads with IgG used as negative control.

Embryo extracts were precleared with empty beads equilibrated in IPP 150 for 30 min at 4°C in the presence of 1× complete protease inhibitor cocktail (Roche). Precleared extracts (300–500 μg) were then added to 60 μl of antibody-bound beads in a final volume of 780 μl of IPP 150 and incubated at 4°C for 4 h on a rotating wheel. Following IP, beads were pelleted, washed four times with five volumes of IPP 150, and RNA extracted using TRIZOL. RNA quality was monitored in a Bioanalyzer. An aliquot of the IP was reserved for protein extraction and IP quality control.

Illumina True-seq libraries were prepared at the CRG Genomics facility using the Ribo-zero kit (Illumina) and subjected to 50-bp single-end sequencing. Reads were aligned to the D. melanogaster ENSEMBL genome release 87 (BDGP release 6_ISO1 MT/dm6 assembly) using the STAR mapper (version 2.5.2a) (Dobin et al, 2013). Qualimap (version 2.2.1) was used to check the quality of the mapped bam files (García-Alcalde et al, 2012). A raw count of reads per gene was obtained with HTSeq (htseq-count function, version 0.6.1p1) (Anders et al, 2015). The R/Bioconductor package DESeq2 was used to assess differential enrichment between experimental samples. Prior to DESeq2, genes for which the sum of raw counts across all samples was less than two were discarded.

Dicer-2 protein interactome identification

Dicer-2 IPs were performed as indicated above, using 300 μg of wild-type (w¹¹¹⁸) or Dicer-2 null extracts, keeping the proportions of extract to beads and volumes. Following IP and washes, one half of the Dicer-2 beads were incubated with 100 U RNase I (Ambion) in 100 μl IPP 150 at 25°C for 15 min on a thermomixer. Untreated samples were incubated in parallel without addition of RNase I. Beads were then washed four times with five vol of IPP 150 before elution of proteins for mass spectrometry and Western blot. For mass spectrometry, samples on beads were washed thrice with 500 μl of 200 mM ammonium bicarbonate (ABC) and resuspended in 60 μl of 6 M urea in 200 mM ABC. Proteins were then reduced by adding DTT (10 μl DTT 10 mM, 37°C, 60 min), alkylated with iodoacetamide (10 μl of IAM 20 mM, 25°C, 30 min), diluted with 200 mM ABC to reach a final urea concentration of 1M, and digested overnight with trypsin (1 μg, 37°C). The peptide mixture was collected and acidified to a final concentration of 5% formic acid. Samples were desalted using a C18 column, evaporated to dryness, and diluted to 10 μl with 0.1% formic acid in milliQ water. Forty-five percent of the peptide mixture was analyzed by LC-MSMS using a 1-h gradient in the LTQ-Orbitrap Velos Pro mass spectrometer (Thermo Fisher Scientific). The data were acquired with Xcalibur software v2.2. and analyzed using the Proteome Discoverer software suite (v1.4.1.14; Thermo Fisher Scientific). The search engine Mascot (v2.5.1 Matrix Science; Perkins et al, 1999) was used for peptide identification. Peptide data were searched against the Uniprot (UP_Drosophila) protein database, and the identified peptides were filtered to a threshold of 5% FDR.

Protein–protein interactions were assessed using SAINTexpress (Teo et al, 2014) and Top3 (Silva et al, 2006) analysis. SAINT interactors with Bayesian false discovery rate of <0.05 were included in the Dicer-2 interactome. For Top3 analysis, the log₂ of the average area of the three most intense peptides of each protein was calculated by Proteome Discoverer. Top3 analysis was constrained by very low abundance or the absence of several proteins in the null IP and by missing peak area values for one of three null mutant replicates. These technical caveats were overcome by semiquantitative scoring of significant peptide occurrences in wild-type versus null replicates and by imputation that substitutes the missing values for the mean value of other replicates, respectively. A t test was performed between the three replicates of each sample condition to identify differentially abundant proteins.

Small scale immunoprecipitation

Protein-A dynabeads (20 μl) were equilibrated in wash buffer (10 mM Hepes pH 8.0, 8% Glycerol) and blocked by incubation with 700 μg embryo extract for 1 h at room temperature. After extensive washing, 2 μl Atx2 serum, 5 μg of affinity-purified αDicer-2 antibody, or IgG were added in wash buffer followed by incubation for 1 h at room temperature and three additional washes. Beads were mixed with 700 μg of embryo extract in 200 μl DE buffer (10 mM Hepes, 5 mM DTT), supplemented with 50 μl of buffer containing 20 mM Hepes, pH 8, 150 mM NaCl, 0.1% NP-40, 1 mM EDTA, 0.5 mM DTT, and 1× protein inhibitor cocktail (Roche) and incubated at 25°C for 1 h on a rotating wheel. Beads were then washed three times with five volumes of 1× NET (50 mM Tris–HCl pH7.5, 150 mM NaCl, 0.1% NP-40, 1 mM EDTA). RNase treatment was performed with 100 U RNase I (Ambion) and 10 μg of RNase A (Sigma-Aldrich) in 40 μl wash buffer with protein inhibitors at room temperature for 20 min. Untreated samples were incubated in parallel without adding RNases. Beads were washed three times with 10 volumes of 1xNET and proteins eluted in SDS buffer for Western blot.

PAT assays

Pools of 100–200 embryos synchronized at 60–90 min were collected in Eppendorf tubes and disrupted with Pellet mixer Cordless (Avantor) in 200 μl Trizol. RNA was obtained using a Maxwell 16 LEV simplyRNA kit (Promega). One microgram of total RNA was used for PAT assays as described in Sallés and Strickland (1995). In some experiments, the poly(A) tail assay kit from Thermo Fisher Scientific (Ref. 764551KT) was used with equivalent results. Gene-specific forward oligonucleotides used for PAT assays were as follows: String (5′-CCGAAACGCAAATGCAAAC-3′), Grapes (5′-CATTGCATTGTTTACGAGTACG-3′), CG8180 (5′-CAGCAGCATTAACTGACGAATCGAC-3′), EDTP (5′-GGCTAAACCGCTCTCCTGTTCTCTAG-3′) and lok (5′-CTGCATTTAAACTGGGCTGCTGCTTC-3′). Amplified products were resolved in agarose gels, and the efficiency of polyadenylation was measured as the ratio of polyadenylated (A+) versus non polyadenylated (A−) RNA, quantified using ImageQuant-TL. The data for Atx2-depleted, Tyf-depleted and Dicer-2 null embryos were normalized to the wt control (set to 100%) that was carried in parallel in each experiment.

RT-qPCR

RNA was reverse-transcribed with SuperScript II (Invitrogen) using a mix of random primers and oligo(dT). The cDNAs were subjected to quantitative real-time PCR using SYBR Green PCR master mix (Applied Biosystems) and ViiA 7 Real-Time PCR System (Applied Biosystems) following the manufacturer′s instructions. Oligonucleotides used in qPCR are as follows: Toll-f (5′-CGCTGCCTTCGCGTCTGTTTGC-3′), Toll-r (5′-GTGTGGAATGCTCGAATAAGTCA-3′), Bicoid-f (5′-ATTGCAATCTGTTAGGCCTCAAG-3′), Bicoid-r (5′-CGGGATCCCGAGTAGAGTAGTTCTTATATATT-3′), Sop-f (5′-CCGTGGTACTGGCATTGTCT-3′), Sop-r (5′-CCGAGTATGCCTGGTAAGGA-3′), Grapes-f (5′-TGAGGAGAATGACCCGATTC-3′), Grapes-r (5′-AACCACCCAGTCTTTCGATG-3′), String-f (5′-CACAAGCGCAACATCATTATC-3′), String-r (5′-CCGGATAGGCGTTGGTATT-3′), CycA-f (5′-GCTGGAGGAGATCACGACTT-3′), CycA-r (5′-TTGTACTTTTCCCGCATGG-3′), CG8180-f (5′-AAGGTCCCAACATCTCTCTGA-3′), CG8180-r (5′-GCTGGGGATGGTATCTCGTA-3′), UbcE2H-f (5′-CAACGATCGCAACAGATCAT-3′), UbcE2H-r (5′-TTTTGCTCTTCCATCCGTTC-3′), RpL32-f (5′-TGCCCACCGGATTCAAGA-3′), RpL32-r (5′-AAACGCGGTTCTGCATGAG-3′), RpS3-f (5′-CGATTTCCAAGAAACGCAAG-3′), RpS3-r (5′-CGAGTCAGGAACTCGTTCAA-3′).