The spliceosome: disorder and dynamics defined

Abstract

Among the many macromolecular machines involved in eukaryotic gene expression, the spliceosome remains one of the most challenging for structural biologists. Defining features of this highly complex apparatus are its excessive number of individual parts, many of which have been evolutionarily selected for regions of structural disorder, and the remarkable compositional and conformation dynamics it must undertake to complete each round of splicing. Here we review recent advances in our understanding of spliceosome structural dynamics stemming from bioinformatics, deep sequencing, high throughput methods for determining protein-protein, protein-RNA and RNA-RNA interaction dynamics, single molecule microscopy and more traditional structural analyses. Together, these tools are rapidly changing our structural appreciation of this remarkably dynamic machine.

Fig 1. An updated spliceosome assembly cycle

A. Structural overview of the yeast spliceosome assembly, activation, and disassembly cycle (adapted from [36]). The crystal structures of human U1 and U4 snRNPs [51,52] and EM structures of larger complexes are shown [6] [53] [54] [55],[56] [57], [58]. B. Shared and unique components of the major and minor spliceosomes. Structures of NTC, U1, U4, and U5 snRNPs are as in A; U11/U12 di-snRNP structure is from [59]. NTC complex association with the minor spliceosome has only been inferred to date, not structurally validated. Also shown are the conserved sequences of major and minor introns with exons (boxes), introns (lettering or solid line), the branch adenosine (bold red) indicated (adapted from[7]).

Fig 2. Intrinsic disorder among spliceosomal proteins

A. Illustration of how IDRs can facilitate formation of larger macromolecular assemblies by adopting ordered structures upon binding to a specific partner and/or serving as sites of post-translational modification. B. Plot showing the correlation between predicted protein disorder of various yeast spliceosomal proteins as calculated by PONDR-FIT (y-axis) and PONDR® VLXT (x-axis) (Reproduced with permission from [21]). These two programs use different algorithms to generate a score for each protein sequence that reflects the fraction of amino acids likely to be in IDRs. This plot shows that both algorithms generate highly correlated outputs with regard to spliceosomal proteins and splicing factors [21]. Using arbitrary PONDRFIT cutoffs, proteins were classified as highly ordered (blue field, 9 proteins), moderately disordered (pink field, 48 proteins) and highly disordered (red field, 52 proteins). Proteins are color coded as indicated to indicate their relationship to different components and complexes.

Fig 3

Ubiquitination and phosphorylation cycles involved in U4/U6.U5 tri-snRNP dynamics (Adapted from [48]). See text for details. Question marks indicate steps that are yet to be elucidated.