-
Katerina R. Katsani,1,
-
Manuel Irimia,2,
-
Christos Karapiperis,3,
-
Zacharias G. Scouras,3,
-
Benjamin J. Blencowe,2,
-
Vasilis J. Promponas4,
Journal name:
Scientific Reports
Volume:
4,
Article number:
4655
DOI:
doi:10.1038/srep04655
Received
04 November 2013
Accepted
-
- Published
There is growing evidence for the involvement of Y-complex nucleoporins (Y-Nups) in cellular processes beyond the inner core of nuclear pores of eukaryotes. To comprehensively assess the range of possible functions of Y-Nups, we delimit their structural and functional properties by high-specificity sequence profiles and tissue-specific expression patterns. Our analysis establishes the presence of Y-Nups across eukaryotes with novel composite domain architectures, supporting new moonlighting functions in DNA repair, RNA processing, signaling and mitotic control. Y-Nups associated with a select subset of the discovered domains are found to be under tight coordinated regulation across diverse human and mouse cell types and tissues, strongly implying that they function in conjunction with the nuclear pore. Collectively, our results unearth an expanded network of Y-Nup interactions, thus supporting the emerging view of the Y-complex as a dynamic protein assembly with diverse functional roles in the cell.
Subject terms:
At a glance
Figures
-
Figure 1: Automated clustering of validated Y-Nup family relationships. Left: Circos Tableviewer (http://mkweb.bcgsc.ca/tableviewer/) representation (see Methods) of the nine automatically generated clusters (C1–C9) and the membership of the detected Y-Nup homologs from iterative sequence profile searches (named in alphabetical order, displayed in counterclockwise fashion). Stripes are color-coded according to clusters (for example, C1 is shown in red, spanning three Y-Nup homologs namely Nup37/SEH1/SEC13); there is a one-to-one correspondence between clusters and homologs except C1 (above), and Nup75 (C6, C7) and Nup107 (C3, C9). Outer circle values represent relative contributions, inner circle values correspond to absolute numbers. Right: Depiction of sequence alignments derived from profile searches (see Data Supplement DS03). Y-Nup families are assigned to the color of their highest-frequency cluster. Full-sized sample alignments are provided for Nup107 (minimum identity 4%, Supplementary Fig. 3) and Nup133 (minimum identity 9%, Supplementary Fig. 4). Note that Nup37 (in C1) and Nup43 (C8) detect fewer homologs due to their restricted phylogenetic distribution.
-
Figure 2: Gene expression patterns of Y-Nups in human (left) and mouse (right) across seven representative tissues based on RNA-seq profiles. Circos Tableviewer (http://mkweb.bcgsc.ca/tableviewer/) representation as in Figure 1. Y-Nups are labeled by their gene names in the corresponding species (as in Figure 1, except NUP85/Nup85 equivalent to Nup75). Tissue labels are self-explanatory (WT signifying wild-type for mouse, -P signifying a single tissue sample). Note that SEC13 exhibits a much higher expression level than SEH1 possibly due to its participation in other macromolecular complexes. The full RNA-seq patterns across a wide range of tissues and cell lines (see Methods) are provided (Data Supplement DS10).
-
Figure 3: A) Multi-domain architecture of representative Y-Nup homologs. Y-Nup domains identified by sequence searches (shown in light blue) found to be associated with protein domains with functions indicating moonlighting roles (shown in green). Grey boxes signify other protein regions. Scale is provided above and below the multi-domain diagram representations. Only the six cases with highest support (Supplementary Table 2) are shown (see also Table 1). A full characterization of the 27 validated domain associations is also provided (Supplementary Fig. 5). The phylogenetic distribution of all 27 reported domains is available in Data Supplement DS08. B) Coordinated RNA-seq expression patterns across twenty representative human tissues between Y-Nups and other domains supported by expression. Pearson correlation coefficient values across the twenty tissues are shown on the x-axis (see Methods). Gene names for human (left) and the encoded labels in this study (right) are shown on the y-axis (Supplementary Table 3), separated by a vertical bar. Full clustering analysis with Spearman rank correlation coefficients supports nine of the eleven cases (green bars) shown here, i.e. all listed genes except TDRD3 and TAF9B (middle block, in Figure 4).
-
Figure 4: Clustering of human gene expression patterns across multiple tissues and cell lines. Gene expression affinities are represented in a heat map by Spearman rank correlation coefficients as similarity measure across tissues and cell lines (see Methods). The narrow strip on both sides of the map indicates the entries corresponding to Y-Nups (red), homologs of associated domains (blue) – including human paralogs (Supplementary Table 3) found in other species (Supplementary Table 2), and a randomly selected set of 300 genes (grey); see also Methods. By splitting the clustering dendrogram at the second bifurcation, three clusters emerge depicted by three major blocks (color-coded, left side). The middle block (light green) is enriched in Y-Nups except SEC13, the upper cluster (light blue) contains SEC13, while the lower cluster (light red) does not contain any known Y-Nups. The inset (right side) magnifies entries in the middle block, where Y-Nups are included: gene names are shown, color scheme as for entire strip; green labels signify genes encoding for Y-Nup associated domains, supported by the bootstrap analysis (Supplementary Table 3). This figure is available at higher resolution as Supplementary Figure 6.
-
Figure 5: An expanded network of Y-Nup interactions from high-throughput experiments and the discovered multi-domain architectures. Using as queries the human gene symbols of the molecules with coordinated gene expression patterns from the Y-Nup enriched middle block (see Figure 4), the resulting network is extracted by GeneMANIA with default parameters (only co-expression, physical and genetic interaction networks are retained). Queries were 60 (of which 3 are not found as interacting), depicted as diamonds (57 in total). An additional 20 genes are discovered by GeneMANIA, depicted as circles. The total genes in this network amount to 77, encompassing a number of known associations of Y-Nups (shown in purple, 8 in number) with other nucleoporins (shown in pink, 4 in number – including Sec13, not in query), e.g. TPR58 and related molecules (shown in light blue, 35 in number), e.g. EXO159. The reported molecules by GeneMANIA include interactions from large-scale experiments not further discussed (shown in grey, 17 in number). The three coordinated gene expression instances regarded as negatives in this study (shown in cyan, 3 in number) are ARG1 (curiously reported by GeneMANIA, thus shown as circle), ACACA and PGM2. The query molecule C2orf34 (synonym: CAMKMT, thus shown as diamond) is also reported by GeneMANIA. The six discovered novel domain associations (shown in purple, 10 in number), include the five of the six molecules with highest support (except CcmE/CycJ, Table 1) and five SMC paralogs (SMC1A, SMC2-4, SMC6), not previously found in association with Y-Nups. GeneMANIA reports no evidence for the association of NUP160-RAD51, NUP98-SET, NUP43-DHX15 and SEH1-TAF9, while providing strong evidence for SEH1-CEP19260, 61, 62. Common genes between PINA & GeneMANIA include other nucleoporins (e.g. NUP93) or others (e.g. EEF1G, Elongation factor 1-y) (see Methods). The annotated layout and GeneMANIA results with supporting literature are available in Data Supplement DS11.
<:footer>
Report this comment #63045
1 – “(figure 2 of cited work)28″ should not be pointing to the manus cript’s figure 2 but Ref 28. Issue only in the HTML version, not the PDF version.
2 – Reference 48 contains, well, a ‘PubMed artifact’ (in PubMed duplicating that title).
Apologies for any inconvenience this might have caused.