Global detection of human variants and isoforms by deep proteome sequencing.

An average shotgun proteomics experiment detects approximately 10,000 human proteins from a single sample. However, individual proteins are typically identified by peptide sequences representing a small fraction of their total amino acids. Hence, an average shotgun experiment fails to distinguish different protein variants and isoforms. Deeper proteome sequencing is therefore required for the global discovery of protein isoforms. Using six different human cell lines, six proteases, deep fractionation and three tandem mass spectrometry fragmentation methods, we identify a million unique peptides from 17,717 protein groups, with a median sequence coverage of approximately 80%. Direct comparison with RNA expression data provides evidence for the translation of most nonsynonymous variants. We have also hypothesized that undetected variants likely arise from mutation-induced protein instability. We further observe comparable detection rates for exon-exon junction peptides representing constitutive and alternative splicing events. Our dataset represents a resource for proteoform discovery and provides direct evidence that most frame-preserving alternatively spliced isoforms are translated.

Autism-Misregulated eIF4G Microexons Control Synaptic Translation and Higher Order Cognitive Functions.

Microexons represent the most highly conserved class of alternative splicing, yet their functions are poorly understood. Here, we focus on closely related neuronal microexons overlapping prion-like domains in the translation initiation factors, eIF4G1 and eIF4G3, the splicing of which is activity dependent and frequently disrupted in autism. CRISPR-Cas9 deletion of these microexons selectively upregulates synaptic proteins that control neuronal activity and plasticity and further triggers a gene expression program mirroring that of activated neurons. Mice lacking the Eif4g1 microexon display social behavior, learning, and memory deficits, accompanied by altered hippocampal synaptic plasticity. We provide evidence that the eIF4G microexons function as a translational brake by causing ribosome stalling, through their propensity to promote the coalescence of cytoplasmic granule components associated with translation repression, including the fragile X mental retardation protein FMRP. The results thus reveal an autism-disrupted mechanism by which alternative splicing specializes neuronal translation to control higher order cognitive functioning.

Multilayered control of exon acquisition permits the emergence of novel forms of regulatory control.

BACKGROUND: The long introns of mammals are pools of evolutionary potential due to the multiplicity of sequences that permit the acquisition of novel exons. However, the permissibility of genes to this type of acquisition and its influence on the evolution of cell regulation is poorly understood. RESULTS: Here, we observe that human genes are highly permissive to the inclusion of novel exonic regions permitting the emergence of novel regulatory features. Our analysis reveals the potential for novel exon acquisition to occur in over 30% of evaluated human genes. Regulatory processes including the rate of splicing efficiency and RNA polymerase II (RNAPII) elongation control this process by modulating the "window of opportunity" for spliceosomal recognition. DNA damage alters this window promoting the inclusion of repeat-derived novel exons that reduce the ribosomal engagement of cell cycle genes. Finally, we demonstrate that the inclusion of novel exons is suppressed in hematological cancer samples and can be reversed by drugs modulating the rate of RNAPII elongation. CONCLUSION: Our work demonstrates that the inclusion of repeat-associated novel intronic regions is a tightly controlled process capable of expanding the regulatory capacity of cells.

Efficient and Accurate Quantitative Profiling of Alternative Splicing Patterns of Any Complexity on a Laptop.

Alternative splicing (AS) is a widespread process underlying the generation of transcriptomic and proteomic diversity and is frequently misregulated in human disease. Accordingly, an important goal of biomedical research is the development of tools capable of comprehensively, accurately, and efficiently profiling AS. Here, we describe Whippet, an easy-to-use RNA-seq analysis method that rapidly-with hardware requirements compatible with a laptop-models and quantifies AS events of any complexity without loss of accuracy. Using an entropic measure of splicing complexity, Whippet reveals that one-third of human protein coding genes produce transcripts with complex AS events involving co-expression of two or more principal splice isoforms. We observe that high-entropy AS events are more prevalent in tumor relative to matched normal tissues and correlate with increased expression of proto-oncogenic splicing factors. Whippet thus affords the rapid and accurate analysis of AS events of any complexity, and as such will facilitate future biomedical research.

Multilayered Control of Alternative Splicing Regulatory Networks by Transcription Factors.

Networks of coordinated alternative splicing (AS) events play critical roles in development and disease. However, a comprehensive knowledge of the factors that regulate these networks is lacking. We describe a high-throughput system for systematically linking trans-acting factors to endogenous RNA regulatory events. Using this system, we identify hundreds of factors associated with diverse regulatory layers that positively or negatively control AS events linked to cell fate. Remarkably, more than one-third of the regulators are transcription factors. Further analyses of the zinc finger protein Zfp871 and BTB/POZ domain transcription factor Nacc1, which regulate neural and stem cell AS programs, respectively, reveal roles in controlling the expression of specific splicing regulators. Surprisingly, these proteins also appear to regulate target AS programs via binding RNA. Our results thus uncover a large "missing cache" of splicing regulators among annotated transcription factors, some of which dually regulate AS through direct and indirect mechanisms.

Molecular Principles of Gene Fusion Mediated Rewiring of Protein Interaction Networks in Cancer.

Gene fusions are common cancer-causing mutations, but the molecular principles by which fusion protein products affect interaction networks and cause disease are not well understood. Here, we perform an integrative analysis of the structural, interactomic, and regulatory properties of thousands of putative fusion proteins. We demonstrate that genes that form fusions (i.e., parent genes) tend to be highly connected hub genes, whose protein products are enriched in structured and disordered interaction-mediating features. Fusion often results in the loss of these parental features and the depletion of regulatory sites such as post-translational modifications. Fusion products disproportionately connect proteins that did not previously interact in the protein interaction network. In this manner, fusion products can escape cellular regulation and constitutively rewire protein interaction networks. We suggest that the deregulation of central, interaction-prone proteins may represent a widespread mechanism by which fusion proteins alter the topology of cellular signaling pathways and promote cancer.

Proteome-wide analysis of human disease mutations in short linear motifs: neglected players in cancer?

Disease mutations are traditionally thought to impair protein functionality by disrupting the folded globular structure of proteins. However, 22% of human disease mutations occur in natively unstructured segments of proteins known as intrinsically disordered regions (IDRs). This therefore implicates defective IDR functionality in various human diseases including cancer. The functionality of IDRs is partly attributable to short linear motifs (SLiMs), but it remains an open question how much defects in SLiMs contribute to human diseases. A proteome-wide comparison of the distribution of missense mutations from disease and non-disease mutation datasets revealed that, in IDRs, disease mutations are more likely to occur within SLiMs than neutral missense mutations. Moreover, compared to neutral missense mutations, disease mutations more frequently impact functionally important residues of SLiMs, cause changes in the physicochemical properties of SLiMs, and disrupt more SLiM-mediated interactions. Analysis of these mutations resulted in a comprehensive list of experimentally validated or predicted SLiMs disrupted in disease. Furthermore, this in-depth analysis suggests that 'prostate cancer pathway' is particularly enriched for proteins with disease-related SLiMs. The contribution of mutations in SLiMs to disease may currently appear small when compared to mutations in globular domains. However, our analysis of mutations in predicted SLiMs suggests that this contribution might be more substantial. Therefore, when analysing the functional impact of mutations on proteins, SLiMs in proteins should not be neglected. Our results suggest that an increased focus on SLiMs in the coming decades will improve our understanding of human diseases and aid in the development of targeted treatments.