Biological impact of mutually exclusive exon switching

Abstract

Alternative splicing can expand the diversity of proteomes. Homologous mutually exclusive exons (MXEs) originate from the same ancestral exon and result in polypeptides with similar structural properties but altered sequence. Why would some genes switch homologous exons and what are their biological impact? Here, we analyse the extent of sequence, structural and functional variability in MXEs and report the first large scale, structure-based analysis of the biological impact of MXE events from different genomes. MXE-specific residues tend to map to single domains, are highly enriched in surface exposed residues and cluster at or near protein functional sites. Thus, MXE events are likely to maintain the protein fold, but alter specificity and selectivity of protein function. This comprehensive resource of MXE events and their annotations is available at: http://gene3d.biochem.ucl.ac.uk/mxemod/. These findings highlight how small, but significant changes at critical positions on a protein surface are exploited in evolution to alter function.

Fig 1. Identification of variable residues and possible functional effects of variable residues.

(A) Identifying the variable residues for an MXE event (MXE-specific residues). The same colour code is used throughout the paper. The MXE region from the pair of proteins generated from the splicing are shown in purple or yellow. Variable residues from the MXE event are shown in red. (B) Possible functional effects of variable residue switching assessed in this paper include altering PPI (Protein-Protein interactions), CSA (catalytic residues from the Catalytic Site Atlas[21]), PSI (Protein-Small molecule Interactions), Allosteric, PTM (Post-Translational Modification) and FunSites (predicted functional sites from functional families (FunFams)).

Fig 2. General statistics of our MXE dataset.

(A) Table showing the number of MXE events and MXE genes in each organism and a matrix with the percentage overlap of MXE gene orthologs between species. The percentage overlap refers to percentages of overlap between gene orthologs between species (e.g. 58% of human MXE genes have an ortholog in mouse). (B) Distributions of MXE exon sizes. MXE sizes were calculated by calculating the length of the MXE. The length of the longer isoforms were plotted. (C) Distribution of sequence identity between MXE pairs identifies one main distribution with its main peak between 50% and 70% sequence revealing likely conserved structure and function. A second much smaller peak around 90% indicates a subset of MXE events that are more recently evolved. Sequence identities were calculated by BLASTing two isoforms against each other. (D) Distribution of number of variable residues between MXE pairs demonstrates that the majority of MXEs have only a limited number of variable residues with almost all events <10 variable residues. Variable residues are amino acid residues that are variable in the alignment between MXE pairs.

Fig 3. Number and names of CATH superfamilies common to multiple species analysed in this study.

MXEs from different species were mapped to CATH domain superfamilies. CATH superfamilies are groups of protein domains with clear evidence of homology. The CATH superfamily code is denoted by four numbers corresponding to each level in the CATH classification (i.e. 3.20.20.120). At the top of the hierarchy is the class level where structural domains are classified based on their secondary structure content. The second level of the hierarchy is the architecture level given by the global arrangement of secondary structures in 3D space. This is followed by the topology level where domains with similar folds (which takes into account the 3D arrangement, orientations and connections between the secondary structures) are grouped together. The fourth level is the homologous superfamily level where domains are deemed homologous. A website tool (http://bioinformatics.psb.ugent.be/webtools/Venn/) was used to draw this Venn diagram.

Fig 4. Structural and functional analysis of the MXE splicing dataset.

(A) The surface exposure of the MXE events compared to random expectation. The first quartile, median, and the third quartile of the population were indicated with lines. NACCESS was used to calculate the relative accessible surface (rASA) of amino acids. Amino acid residues were considered to be exposed if the rASA value was above 10%. (B) The surface exposure of the MXE variable residues compared to the MXE non-variable residues. We compared the solvent exposure of the variable residues versus the solvent exposure of the MXE conserved residues for each MXE event. NACCESS was used to calculate the relative accessible surface (rASA) of amino acids. Amino acid residues were considered to be exposed if the rASA value was above 10%. (C) Distribution of McLachlan scores for the variable residues. For each MXE, we summed up all the McLachlan similarity scores for a set of variable residues and divided by the total number of mapped variable residues to normalise the score. (D) Proximity of MXE variable residue clusters to amino acids assigned to different functional classes (indicated by Icons) CSA = catalytic residues from the Catalytic Site Atlas, PPI = protein interaction sites and PSI = protein-small molecule interactions). A breakdown of this Fig by individual species is available in S10 Fig. We used the Z-score test to compute the statistical significance. Where appropriate the level of statistical significance is shown above each type of functional site.

Fig 5. Examples of MXE events affecting amongst other things.

(A) Protein-protein interaction of the PKM gene. Cancer mutated residues are shown in spacefill. (B) Allosteric regulation of the PFKP gene. Cancer mutated residues are shown in spacefill. (C) Small molecule interaction (PSI) sites in the CRMP gene. We are only showing variable binding zinc/catalytic residues. Zinc molecules are shown green spacefill. (D) Catalytic site residues of gene CG42249 from drosophila. We are only showing variable binding zinc/catalytic residues.

Fig 6. MXE-MOD Website.

(A) The home page of MXE-MOD website. (B) The browser view allows the user to search different organisms and find a particular gene of interest. Annotations for the gene and associated MXE event have been integrated from different public resources: CATH, GENEONTOLOGY, Human Protein Atlas, DRUGBANK and Ensembl [,–49].(C-D) Detailed ‘MXE model’ pages for 2 different examples in the website. We can see the alternative MXE structure superposed in (C). We can see the variable PPI residues shown in space fill in the website in (D). MXE-MOD utilises the PV javascript viewer [50]. For both examples we show the expression patterns for the MXE isoform groups at different stages of development from RNA-seq and proteomics derived datasets. RNA-seq data is obtained from ModEncode resource [51] and proteomics data from the DDIP consortium (https://ddip-proteome.org) led by Simon Hubbard, Manchester University, UK.