Fig 2: Translational repression alone underlies many of the downstream molecular changes associated with miRNA loss.(A) Comparison between mRNA stability changes in Dgcr8 KO versus Ddx6 KO cells. n = 3 for wild-type, n = 4 for Ddx6 KO (2 replicates of each Ddx6 KO line), n = 3 for Dgcr8 KO. (B) Comparison between translation level changes in Dgcr8 KO versus Ddx6 KO cells. n = 3 for each genotype. (C) Comparison between mRNA changes in Dgcr8 KO versus Ddx6 KO cells. The p value was calculated with correlation significance test. (D) Summary schematic comparing Dgcr8 KO cells to Ddx6 KO cells. Dgcr8 KO leads to the loss of both translational repression and mRNA destabilization of miRNA targets, while Ddx6 KO only leads to the loss of translational repression of miRNA targets. mRNA stability is measured as the ratio of mRNA/4sU reads, changes in translation level are measured as the ratio of polysome/monosome reads, protein level changes are not directly measured but are predicted based on mRNA stability and translation level changes. Changes in translation level alone in Ddx6 KO cells produce similar phenotypes and global molecular changes to Dgcr8 KO cells. See also Figure 5—figure supplement 1.

Fig 3: DGCR8 and the Exosome Coexist in a Complex(A) Sedimentation patterns of immunopurified FLAG-Drosha and FLAG-DGCR8 native complexes in 5%–30% glycerol gradient fractions, as revealed by western blot analysis with an anti-FLAG antibody. “Light” denotes lighter-molecular-weight fractions, whereas “heavy” indicates heavier molecular fractions. The migration of the molecular weight markers is indicated at the top (to see uncropped versions of these images, see Figure S2B).(B) Western blot of coimmunoprecipitated hRRP6 with FLAG-Drosha (top panel) and FLAG-DGCR8 (bottom panel) after glycerol gradient fractionation. Fractions from a 5%–30% glycerol gradient were pooled into light (lane 2), corresponding to fractions 1–11, and heavy (lane 3), corresponding to fractions 12–22, and run in a single lane for sensitivity purposes.(C) Sedimentation patterns of endogenous Drosha, DGCR8, hRRP6, and hRRP41 proteins in 5%–30% glycerol gradients from nuclear HEK293T cell extracts, as revealed by western blot analysis with specific antibodies. Lysates run in all gradients were produced in the presence of DNase and RNase.

Fig 4: DGCR8/hRRP6 Complex Controls Human Telomerase RNA Levels(A) Distribution of DGCR8 and hRRP6 CLIP reads over hTR loci; numbers on the left represent number of reads obtained from each library mapping to hTR.(B and C) Northern analyses of associated hTR RNA with immunoprecipitated endogenous DGCR8 (lane 3) and hRRP6 (lane 4) in HEK293T cells (B) and with mouse RRP6 in the presence (Dgcr8+/+, lane 3) and absence (Dgcr8−/−, lane 5) of DGCR8 in mESC (C).(D) HeLa cells were transiently depleted of DGCR8, hRRP6 and Drosha and hTR levels were quantified by qRT-PCR (for depletion levels, see Figures S5D and S7B).(E) Levels of mouse TERC RNA were quantified by qRT-PCR in the absence of DGCR8 (Dgcr8−/−) and Dicer (Dicer−/−). All values represented in panels (D) and (E) are the average of at least three biological replicates ± SEM. Asterisks denote siginificant p value (≤0.05) by Student’s t test.(F) Relative telomere length quantification by qPCR of genomic DNA from cells lacking DGCR8 (Dgcr8−/−) and Dicer (Dicer−/−) and their respective wild-type controls (Dgcr8+/+ and Dicer+/+). Numbers in brackets represent the passage number. Absolute telomere quantification was normalized to a single-copy gene (c-myc), as described (Callicott and Womack, 2006). Values represent the average of three biological replicates ± SD.

Fig 5: Depletion of DGCR8 and hRRP6 Specifically Stabilizes Mature snoRNAs(A) Schematic representation of U16 snoRNA location in intron 3 of the host RPL4 pre-mRNA. BS, branch site; E4, exon 4.(B) HeLa cells were transiently depleted of DGCR8, hRRP6, hDIS3, RBM7, and ZCCHC7, and the levels of mature U16 were quantitated by qRT-PCR, using primers depicted on top of the panel.(C) Quantification of the host pre-mRNAs containing U16 snoRNA in HeLa cells depleted for all factors depicted in (B), but also including hRRP41. For levels of depletion in (B) and (C), see Figure S5D. All values represented in the two panels are the average of at least three biological replicas showing ± SEM. Asterisks denote significant p value (=0.05) by Student’s t test.