High-throughput analyses of hnRNP H1 dissects its multi-functional aspect - PubMed (original) (raw)

High-throughput analyses of hnRNP H1 dissects its multi-functional aspect

Philip J Uren et al. RNA Biol. 2016.

Abstract

hnRNPs are polyvalent RNA binding proteins that have been implicated in a range of regulatory roles including splicing, mRNA decay, translation, and miRNA metabolism. A variety of genome wide studies have taken advantage of methods like CLIP and RIP to identify the targets and binding sites of RNA binding proteins. However, due to the complex nature of RNA-binding proteins, these studies are incomplete without assays that characterize the impact of RBP binding on mRNA target expression. Here we used a suite of high-throughput approaches (RIP-Seq, iCLIP, RNA-Seq and shotgun proteomics) to provide a comprehensive view of hnRNP H1s ensemble of targets and its role in splicing, mRNA decay, and translation. The combination of RIP-Seq and iCLIP allowed us to identify a set of 1,086 high confidence target transcripts. Binding site motif analysis of these targets suggests the TGGG tetramer as a prevalent component of hnRNP H1 binding motif, with particular enrichment around intronic hnRNP H1 sites. Our analysis of the target transcripts and binding sites indicates that hnRNP H1s involvement in splicing is 2-fold: it directly affects a substantial number of splicing events, but also regulates the expression of major components of the splicing machinery and other RBPs with known roles in splicing regulation. The identified mRNA targets displayed function enrichment in MAPK signaling and ubiquitin mediated proteolysis, which might be main routes by which hnRNP H1 promotes tumorigenesis.

Keywords: Integrated analysis; Proteomics; RIP-seq; RNA-binding proteins; RNA-seq; hnRNP H1; iCLIP.

PubMed Disclaimer

Figures

Figure 1.

Characterization of the hnRNP H1 binding site. (A) Autoradiograph showing radioactively labeled hnRNP H1 protein–RNA complexes after immunoprecipitation using anti-hnRNP H1 antibody. Normal rabbit IgG was used as negative control. (B) The top motifs identified by de novo motif finding around hnRNP H1 iCLIP sites in 3′UTR, 5′UTR, coding-sequence and intronic contexts (C) The observed/expected ratio of triple-G occurrences around hnRNP H1 iCLIP sites. Expected counts were estimated from a set iCLIP datasets for other RNA binding proteins (see methods). (D) The observed/expected ratio of trimers around hnRNP H1 iCLIP sites. Expected counts computed as in C. (E) Probability of RNA adopting single-stranded conformation around hnRNP H1 iCLIP sites.

Figure 2.

Heterogeneous nuclear ribonucleoprotein H1 is a master controller in a feed-forward network of splicing regulation. hnRNP H1 identified targets and significantly enriched KEGG pathways and biological processes (corrected p < 0.01). Components of the most enriched pathway (the spliceosome) are shown in the center and highlighted in yellow.

Figure 3.

Impact of hnRNP H1 on mRNA stability. (A) Log fold-change and corrected p-values for all genes with significant changes in mRNA levels, colored by whether they were identified as direct targets of hnRNP H1. (B) The proportion of mRNA-level changes that are up- or down-regulation for each decile of FDR scores. High confidence changes are preferentially found among genes displaying downregulation upon hnRNP H1 knockdown. The same trend is observed for direct and indirect targets. (C) Corrected p-values for up- and down-regulated genes (direct and indirect targets) upon hnRNP H1 knockdown; target genes showed more significant changes in both up- and down-regulation upon hnRNP H1 knockdown. Outliers removed for visualization only. (D) The genes showing significant differential expression upon H1 knockdown significantly overlap those genes that are differentially expressed when transcription is stalled (** p< 0.001, Wilcoxon rank sum test, *** p < 0.001, Fisher's exact test).

Figure 4.

Impact of hnRNP H1 on RNA splicing. (A) Log fold-change and corrected p-values for all exons with significant changes in inclusion rates, colored by whether they were identified as targets of hnRNP H1. (B) The proportion of changes in exon-inclusion rate that are increases or decreases. Results are split by whether the exon is in a target of hnRNP H1 or not. (C) The proportions of exons, which show significant increased and decreased usage within hnRNP H1 targets that are in 3′ UTR, 5′ UTR or coding exons. (D) Frequency of hnRNP H1 iCLIP sites at exon/intron boundaries for exons showing up-regulation upon hnRNP H1 knockdown and those showing downregulation (E) Examples of increased exon inclusion and exclusion upon knockdown of hnRNPH1 respectively (from left to right). Highlighted are regions of strong RIP-Seq and iCLIP binding activity.

Figure 5.

Impact of hnRNP H1 on intron retention. (A) Volcano plot showing the significance and strength of changes in intron levels for the 2727 introns with significant changes. There is little preference toward either down- or up- regulation on H1 knockdown. (B) The proportion of significantly changed exons that show increased inclusion versus decreased inclusion on hnRNP H1 knockdown based on whether they are putative iCLIP targets or not - here we consider an intron a putative target if it contains a significant iCLIP site in the intron body or either of the flaking exons. (C) The absolute number and relative proportions of exons showing up- or down-regulation on hnRNP H1 knockdown dependent on iCLIP target type. Those marked 3′ exon, 5′ exon or intron contain iCLIP sites only in those regions, while ‘multiple’ refers to exons with iCLIP sites in more than one of those regions.

Figure 6.

Impact of hnRNP H1 on alternative polyadenylation. (A) The number of genes showing evidence of lengthened or shortened 3′ UTRs on hnRNP H1 knockdown and the proportion of each that have iCLIP sites in their 3′ UTR and were identified as RIP targets. (B) Changes in poly-adenylation sorted by confidence of change shows enrichment of putative targets in high-confidence changes. (C) Profile of normalized iCLIP site count around poly-A tail locations shows enrichment of binding in the proximal 5′ area. (D) qRT-PCR results, validating the effect of hnRNP H1 on the selection of poly-adenylation site for a set of predicted targets (**p < 0.01).

Figure 7.

Changes in protein abundance upon hnRNP H1 knockdown. (A) Fold-change in mRNA levels vs. fold-change in protein concentrations for those transcripts where significant changes were detected in both data sets, stratified by whether the transcript was identified as a putative hnRNP H1 target or not. Shown above is Spearman's rank correlation coefficient for changes in putative targets and non-targets. Both correlations are significant (***p < 0.001), but stronger in the case of putative targets. Error-bars are at the 95% confidence interval. (B) Distribution of iCLIP sites within genes that show up- or downregulation of mRNA levels, protein levels, or both upon hnRNP H1 knockdown. (C) Decreased expression of ATNX10, CTH and RBM3 after hnRNP H1 knockdown as evaluated by immunoblotting. Each experiment was performed 3 times and tubulin was used as a loading control in each western analysis.

References

1. Gerstberger S, Hafner M, Tuschl T. A census of human RNA-binding proteins. Nature reviews Genetics 2014; 15(12):829-845; PMID:25365966; http://dx.doi.org/10.1038/nrg381322574288 - PMC - PubMed
1. Huelga SC, Vu AQ, Arnold JD, Liang TY, Liu PP, Yan BY, Donohue JP, Shiue L, Hoon S, Brenner S, et al.. Integrative genome-wide analysis reveals cooperative regulation of alternative splicing by hnrnp proteins. Cell Rep 2012; 1(2):167-178; PMID:22574288; http://dx.doi.org/10.1016/j.celrep.2012.02.001 - DOI - PMC - PubMed
1. Han S, Tang Y, Smith R. Functional diversity of the hnRNPs: past, present and perspectives. Biochem J 2010; 430:379-392; PMID:20795951; http://dx.doi.org/10.1042/BJ20100396 - DOI - PubMed
1. Busch A, Hertel KJ. Evolution of SR protein and hnRNP splicing regulatory factors. Wiley Interdiscipl Rev: RNA 2012; 3(1):1-12; PMID:21898828; http://dx.doi.org/10.1002/wrna.100 - DOI - PMC - PubMed
1. Bent H, Hanne HR, Henrik V, Kurt D, Xiangdang L, Pavel G, Peder M, Borbala G, Niels T, Julio EC.. Heterogeneous nuclear ribonucleoproteins H, H', and F are members of a ubiquitously expressed subfamily of related but distinct proteins encoded by genes mapping to different chromosomes. J Biol Chem 1995; 270(48):28780-28789; PMID: 7499401; http://dx.doi.org/2105749610.1074/jbc.270.48.28780 - DOI - PubMed

High-throughput analyses of hnRNP H1 dissects its multi-functional aspect - PubMed (original) (raw)