Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2014 Jan;21(1):26-35.
doi: 10.1038/nsmb.2739. Epub 2013 Dec 15.

Staufen1 Senses Overall Transcript Secondary Structure to Regulate Translation

Affiliations
Free PMC article

Staufen1 Senses Overall Transcript Secondary Structure to Regulate Translation

Emiliano P Ricci et al. Nat Struct Mol Biol. .
Free PMC article

Abstract

Human Staufen1 (Stau1) is a double-stranded RNA (dsRNA)-binding protein implicated in multiple post-transcriptional gene-regulatory processes. Here we combined RNA immunoprecipitation in tandem (RIPiT) with RNase footprinting, formaldehyde cross-linking, sonication-mediated RNA fragmentation and deep sequencing to map Staufen1-binding sites transcriptome wide. We find that Stau1 binds complex secondary structures containing multiple short helices, many of which are formed by inverted Alu elements in annotated 3' untranslated regions (UTRs) or in 'strongly distal' 3' UTRs. Stau1 also interacts with actively translating ribosomes and with mRNA coding sequences (CDSs) and 3' UTRs in proportion to their GC content and propensity to form internal secondary structure. On mRNAs with high CDS GC content, higher Stau1 levels lead to greater ribosome densities, thus suggesting a general role for Stau1 in modulating translation elongation through structured CDS regions. Our results also indicate that Stau1 regulates translation of transcription-regulatory proteins.

Figures

Figure 1
Figure 1
Mapping of Stau1 RNA-binding sites reveals coding regions and 3′ UTRs as major occupancy sites. (a) Scheme of the tagged WT and mut Stau1 proteins used in this study. (b) Scheme of tandem affinity purifications for footprinting (FOOT) and cross-linking (CROSS) library construction. (c) Pie charts showing the distribution of sequencing reads for each library. (d) Example of Stau1 cross-inking signal across the CDS of ALDOA (NM_184041.2). (e) Composite plot of the distribution of sequencing reads across the 5′ UTR, CDS and 3′ UTR of all genes for RNA-seq (red), Stau1-WT CROSS (blue) and Stau1-mut CROSS (black) libraries. (f) Per-gene scatter plot of CDS ribo-seq read density versus CDS Stau1-WT CROSS read density with the associated Spearman correlation and calculated P value (n = 2 biological replicates).
Figure 2
Figure 2
Inverted Alu pairs are an important class of Stau1-binding sites. (a,b) Distribution of sequencing reads obtained from RNA-seq (green), Stau1-WT CROSS (yellow), Stau1-mut CROSS (blue), Stau1-WT FOOT (brown), Stau1-mut FOOT (violet) and PAS-seq libraries (black) for the 3′ UTR of PAICS (NM_001079524) (a) and the strongly distal 3′ UTR of BRI3BP (NM_080626.5) (b). (c,d) Per-gene scatter plots of Stau1-WT CROSS and Stau1-mut CROSS read counts under called 3′-UTR Stau1-WT CROSS peak positions with associated Spearman correlation and calculated P values (n = 2 biological replicates). Genes containing a 3′ UTR Alu pair are colored with respect to the distance between each tandem Alu pair (c) or inverted Alu pair (d). The dashed line corresponds to the 2.7 cutoff in the ratio of Stau1-WT over mut read counts.
Figure 3
Figure 3
Characterization of the structural features of Stau1 Alu-binding sites. (a) Composite plot of Stau1-WT CROSS, Stau1-mut CROSS and RNA-seq read counts around tandem or inverted Alu pairs. Read counts normalized to host gene reads per kilobase per million mapped reads (RPKM) were determined for a region spanning 1 kb up- and downstream of the paired Alu elements separated by the indicated distance (n = 2 biological replicates). (b) Example of a 3′-UTR inverted Alu-binding site in C11orf58 (NM_014267.5) showing read counts per million mapped (pmm) reads for RNA-seq, Stau1-WT CROSS and Stau1-mut CROSS libraries. The centroid secondary structure for this Alu pair predicted with the Vienna folding package is shown below center. (c,d) Length distribution of predicted helices (c) and loop (d) regions within secondary structures of 3′-UTR inverted Alu pairs or 3′-UTR sequences of identical size randomly picked from nontarget genes. P values corresponding to the comparison of helix and loop length distributions between Stau1 Alu targets and random 3′ UTRs were calculated with the Wilcoxon rank-sum test (n = 2 biological replicates).
Figure 4
Figure 4
Examples of 3′-UTR non-Alu Staufen-binding sites. (a–d) Read distributions for RNA-seq (green), Stau1-WT CROSS (yellow) and Stau1-WT FOOT (brown) libraries on the 3′ UTRs of LMBR1 (NM_022458.3) (a), TEP1 (NM_007110.4) (b), IGF2BP1 (NM_006546.3) (c) and MDM2 (NM_002392) (d) (left) together with the corresponding centroid secondary structure colored for base-pairing probability as predicted by the Vienna folding package (right). Numbers below the Stau1 WT FOOT track and in the predicted secondary structure correspond to Stau1-binding sites.
Figure 5
Figure 5
Stau1 binding to 3′ UTRs correlates with GC content and predicted secondary-structure free energy. (a) Per-gene scatter plot of 3′-UTR predicted secondary-structure free energy normalized by the length of the 3′ UTR (kcal/mol/nucleotide) against Stau1 WT/mut 3′-UTR ratio (log2). (b) Per-gene scatter plot of 3′-UTR GC content (%) against Stau1 WT/mut 3′-UTR ratio (log2) with associated Spearman correlation and calculated P values. Red and yellow dots correspond to called Alu and non-Alu binding sites, respectively (n = 2 biological replicates).
Figure 6
Figure 6
Stau1 occupancy on the CDS strongly correlates with GC content and predicted secondary-structure free energy. (a) Composite plot of the distribution of sequencing reads across the 5′ UTR, CDS and 3′ UTR of called Stau1-target genes for RNA-seq (red), Stau1-WT CROSS (blue) and Stau1-mut CROSS (black) ibraries. (b) Per-gene scatter plot (log10) of total CDS Stau1-WT CROSS read counts/tota CDS Stau1-mut CROSS reads counts (CDS ratio) versus the ratio of Stau1-WT CROSS and Stau1-mut CROSS read counts under called 3′-UTR Stau1-WT CROSS peak positions (3′-UTR peak ratio). Red and yellow dots correspond to called Alu- and non-Alu–binding sites, respectively. (c) Per-gene scatter plot of 3′ UTR against CDS predicted secondary-structure free energy normalized by the length of the 3′ UTR (kcal/mol/nucleotide). (d) Per-gene scatter plot of 3′ UTR against CDS GC content (%). (e) Per-gene scatter plot of CDS predicted secondary-structure free energy normalized by the length of the CDS (kcal/mol/nucleotide) against Stau1 WT/mut 3′-UTR ratio (log2). (f) Per-gene scatter plot of CDS GC content (%) against Stau1 WT/mut 3′-UTR ratio (log2). All correlation coefficients and P values were calculated with the Spearman rank correlation (n = 2 biological replicates) throughout figure.
Figure 7
Figure 7
Consequences of Stau1 binding on RNA levels and ribosome density. (a) Inset, western blot of lysates from control cells, cells expressing anti-Stau1 shRNA and cells overexpressing Flag–Stau1-WT. Bar graph, quantitation of inset blot (n = 2). Uncropped gel image is shown in Supplementary Figure 7a. (b) Workflow for ribosome profiling and RNA-seq analysis for cells expressing anti-Stau1 shRNA or overexpressing Flag–Stau1-WT. (c) Cumulative plots of cytoplasmic RNA levels (left) and ribosome density (right) fold change (log2) between cells overexpressing Flag–Stau1-WT and cells expressing anti-Stau1 shRNA, based on Stau1 WT/mut CDS ratio. (d) Same as c but based on CDS GC content. (e) Box-plot representation of mRNA levels (left) and translation efficiency (right) fold change (log2 scale) between cells overexpressing Flag–Stau1-WT and cells expressing anti-Stau1 shRNA for genes lacking Stau1 3′ UTR–binding sites (nontargets), genes with 3′-UTR inverted Alu target sites (Stau1 Alu targets) and genes with 3′-UTR non-Alu Stau1-target sites (Stau1 non-Alu targets). The upper and lower ‘hinges’ correspond to the first and third quartiles. The upper whisker extends from the hinge to the highest value within 1.5× interquartile range (or distance between the first and third quartiles) of the hinge. The lowest whisker extends from the hinge to the lowest value within 1.5× interquartile range of the hinge. Data points beyond the end of the whisker correspond to outliers. All P values were calculated with the Kolmogorov–Smirnov test (n = 2 biological replicates) throughout figure.
Figure 8
Figure 8
Model of Stau1 RNA binding and its functional role in translation. Stau1 interacts with both actively translating ribosomes and secondary structures in CDS and 3′-UTR regions. Some 3′ UTRs contain highly complex secondary structures (for example, inverted Alu pairs) that serve as kinetically stable Stau1-binding sites. However, Stau1 also makes transient interactions with smaller secondary structures throughout CDS and 3′-UTR regions. Formation of these structures is a function of overall CDS and 3′-UTR GC content. Whereas interaction of Stau1 with 3′-UTR Alu pairs has a small positive effect on cytoplasmic mRNA levels, high Stau1 CDS occupancy both increases ribosome density and slightly decreases cytoplasmic mRNA levels.

Similar articles

See all similar articles

Cited by 61 articles

See all "Cited by" articles

Publication types

Associated data

Feedback