Extensive cargo identification reveals distinct biological roles of the 12 importin pathways

Abstract

Vast numbers of proteins are transported into and out of the nuclei by approximately 20 species of importin-β family nucleocytoplasmic transport receptors. However, the significance of the multiple parallel transport pathways that the receptors constitute is poorly understood because only limited numbers of cargo proteins have been reported. Here, we identified cargo proteins specific to the 12 species of human import receptors with a high-throughput method that employs stable isotope labeling with amino acids in cell culture, an in vitro reconstituted transport system, and quantitative mass spectrometry. The identified cargoes illuminated the manner of cargo allocation to the receptors. The redundancies of the receptors vary widely depending on the cargo protein. Cargoes of the same receptor are functionally related to one another, and the predominant protein groups in the cargo cohorts differ among the receptors. Thus, the receptors are linked to distinct biological processes by the nature of their cargoes.

Keywords: SILAC; biochemistry; cell biology; human; importin; nuclear protein; nuclear transport; transportin.

Figure 1.. SILAC-Tp effectively sorts Trn-1 cargoes.

(A) Z-scores in the Trn-1 2nd- and 3rd-Z-rankings. The second (left) and third (right) Z-scores are presented for the top 250 proteins in the Trn-1 2nd- and 3rd-Z-rankings, respectively. The total number of ranked (quantified) proteins and the number of previously reported cargoes included in the ranking are indicated at the bottom. The magenta bars represent previously reported cargoes. The blue and dark gray bars represent the proteins that did and did not bind directly to Trn-1, respectively, in the bead halo assays (Supplementary file 2). Identical proteins marked by the colors are connected by lines. Proteins that carry PY-NLS motifs are indicated by green bars. (B) Distribution of PY-NLS motif-containing proteins in the rankings. The percentage of the proteins carrying PY-NLS motifs in 50 consecutively aligned proteins is presented along with the 2nd- and 3rd-Z-rankings (left and right, respectively). For example, the top 50 proteins in the 2nd-Z-ranking include 19 (38%) PY-NLS motif-containing proteins, and thus the value at position 1 is 38%. Two types of PY-NLS motifs, basic and hydrophobic, are defined as presented at the bottom. DOI: http://dx.doi.org/10.7554/eLife.21184.002

Figure 1—figure supplement 1.. SILAC-Tp experimental system.

(A) Methods outline. For details, see the Materials and methods. This figure was originally published by Kimura et al. (2013a). (B) Depletion of the Imp-β family NTRs from the cytosolic and nuclear extracts. The cytosolic and nuclear extracts before and after the NTR depletion were analyzed by Western blotting with antibodies specific for each NTR. Asterisk, no band was detected. For Imp-11, the lower band indicated by the arrowhead was removed. For the antibodies, see Supplementary file 12B. (C) In vitro transport of specific cargoes. The reported cargoes were expressed as GFP-fusion proteins and purified. Then, the cargoes were added to the in vitro transport system with (+) or without (–) each specific NTR. The import was observed by fluorescence microscopy. IBB, Imp-β-binding domain of Imp-α. For the protein accessions, see Supplementary file 12B. DOI: http://dx.doi.org/10.7554/eLife.21184.004

Figure 1—figure supplement 2.. Trn-1 cargoes are effectively sorted by the second or third Z-scores in three replicates of SILAC-Tp.

(A) Frequency distributions of the log₂[(L/H_+NTR)/(L/H_Ctl)] = log₂(+NTR/Ctl) in three replicates of SILAC-Tp with Trn-1. (B) Frequency distributions of the Z-scores of log₂[(L/H_+NTR)/(L/H_Ctl)] in three replicates of SILAC-Tp with Trn-1. The Z-score of each protein was calculated as Z=(X-μ)/σ, where X is log₂[(L/H_+NTR)/(L/H_Ctl)] of a protein, μ is the mean of X in one replicate, and σ is the standard deviation of X. (C) Distribution of the three Z-scores in the Trn-1 2nd-Z-ranking. The Trn-1 2nd-Z-ranking consists of 1649 proteins. Both of the two or all three of the Z-scores are plotted against the rank by the second Z-score. The Z-scores in one replicate are presented as circles with the same color (blue, green, or violet), and the diamonds drawn in dark blue, dark green, and dark violet represent the Z-scores of reported cargoes in the respective replicates. (D) Distribution of the three Z-scores in the Trn-1 3rd-Z-ranking. The Trn-1 3rd-Z-ranking consists of 1235 proteins. All three of the Z-scores are plotted against the rank by the third Z-score. The marker colors are the same as those in (C). (E) Z-scores of the top 200 proteins in the Trn-1 2nd-Z-ranking. Close up of the top 200 in (C). (F) Z-scores of the top 200 proteins in the Trn-1 3rd-Z-ranking. Close up of the top 200 in (D). DOI: http://dx.doi.org/10.7554/eLife.21184.005

Figure 1—figure supplement 3.. Reported cargo rates and recall of the Trn-1 2nd- and 3rd-Z-ranking.

(A) Rates of reported cargoes in the Trn-1 2nd- and 3rd-Z-percentile rankings. The reported cargo rates in the top 1% to 50% ranks are shown under the two definitions of negative examples: unreported, proteins not reported as cargoes; and non-nuclear, proteins annotated with non-nuclear localization. (B) Recall by ranks of the Trn-1 2nd- and 3rd-Z-percentile rankings. The recall of reported cargoes in the top 1% to 50% ranks are indicated. See also Figure 1—source data 1. DOI: http://dx.doi.org/10.7554/eLife.21184.006

Figure 2.. Trn-1, Imp-13, and Trn-SR cargo rankings.

(A–C) The top 100 proteins in the Trn-1 (A), Imp-13 (B), and Trn-SR (C) 2nd- and 3rd-Z-rankings (left and right, respectively). Magenta, reported cargoes; blue, proteins bound directly to the NTR in the bead halo assays (Supplementary file 2); orange in (C), SR-rich SFs that have not been reported; and green in (C), other RS (SR)-domain proteins. Identical proteins marked by the colors are connected by lines. DOI: http://dx.doi.org/10.7554/eLife.21184.007

Figure 2—figure supplement 1.. Imp-13 cargoes are effectively sorted by the second or third Z-scores in three replicates of SILAC-Tp.

(A) Distribution of the three Z-scores in the Imp-13 2nd-Z-ranking. The Imp-13 2nd-Z-ranking consists of 2060 proteins. Both of the two or all three of the Z-scores are plotted against the rank by the second Z-score. The Z-scores in one replicate are presented as circles with the same color (blue, green, or violet), and the diamonds drawn in dark blue, dark green, and dark violet represent the Z-scores of reported cargoes in the respective replicates. Z-scores of the top 200 proteins are presented. (B) Distribution of the three Z-scores in the Imp-13 3rd-Z-ranking. The Imp-13 3rd-Z-ranking consists of 1671 proteins. All three of the Z-scores are plotted against the rank by the third Z-score. The marker colors are the same as those in (A). Z-scores of the top 200 proteins are presented. DOI: http://dx.doi.org/10.7554/eLife.21184.008

Figure 2—figure supplement 2.. SILAC-Tp effectively sorts Imp-13 and Trn-SR cargoes.

(A) Z-scores in the Imp-13 2nd- and 3rd-Z-rankings. The second (left) and third (right) Z-scores are presented for the top 250 proteins in the Imp-13 2nd- and 3rd-Z-rankings, respectively. The total number of the ranked (quantified) proteins and the number of the previously reported cargoes included in the ranking are indicated at the bottom. The magenta bars represent the reported cargoes. The blue and dark gray bars represent proteins that did and did not bind directly to Imp-13, respectively, in the bead halo assays (Supplementary file 2). Identical proteins marked by the colors are connected by lines. (B) Z-scores in the Trn-SR 2nd- and 3rd-Z-rankings. The Z-scores in the Trn-SR 2nd- and 3rd-Z-rankings are presented as in (A). The orange and green bars represent SR-rich splicing factors that have not been reported and other RS-domain proteins, respectively. DOI: http://dx.doi.org/10.7554/eLife.21184.009

Figure 2—figure supplement 3.. Imp-β cargo ranking and Z-scores in the 2nd- and 3rd-Z-rankings.

(A) The top 100 proteins in the Imp-β 2nd- and 3rd-Z-rankings (left and right, respectively). Blue indicates proteins bound directly to Imp-β or both Imp-α and -β, and light blue indicates proteins bound directly to Imp-α but not Imp-β in the bead halo assays (Supplementary file 2). Yellow highlight indicates Imp-α. Identical proteins marked by the colors are connected by lines. (B) Z-scores in the Imp-β 2nd- and 3rd-Z-rankings. The second (left) and third (right) Z-scores are presented for the top 250 proteins in the Imp-β 2nd- and 3rd-Z-rankings, respectively. The total number of the ranked proteins and the number of previously reported Imp-β direct cargoes that are included in the ranking are indicated at the bottom. Bar colors: blue, proteins bound directly to Imp-β or both Imp-α and -β in the bead halo assays; light blue, proteins bound directly to Imp-α but not Imp-β; dark gray; proteins that did not bind to either Imp-α or Imp-β; and yellow, Imp-α. DOI: http://dx.doi.org/10.7554/eLife.21184.010

Figure 3.. 3rd-Z-4% cargoes of the 12 NTRs.

The 3rd-Z-4% cargoes of each NTR are listed by the gene names in the 3rd-Z-rank orders. The ranks by the second Z-scores are also shown. The 3rd-Z-4% and 2nd-Z-15% cargoes are indicated by cyan in the rank columns. Colors in the gene name columns: magenta, reported cargoes; blue, cargoes bound directly to the NTR in the bead halo assays (Supplementary file 2); light blue, cargoes bond directly to Imp-α but not to Imp-β; gray, proteins that did not bind to the NTRs; and yellow, Imp-α. For the 2nd-Z-15% cargoes, see Supplementary file 3. DOI: http://dx.doi.org/10.7554/eLife.21184.011

Figure 4.. Phylogenetic tree and cargo profile hierarchical clustering of the Imp-β family import receptors.

(A) Phylogenetic tree of the 12 Imp-β family import receptors with the bootstrap values. Scale bar indicates substitutions per site. (B) A hierarchical clustering dendrogram of the same NTRs (except Imp-β) based on the similarities of their 3rd-Z-4% cargo profiles. Imp-β was excluded because Imp-α connects to Imp-β and many of the identified cargoes. The scale indicates the intercluster distance. (C) The numbers of 3rd-Z-4% cargoes shared by two NTRs. For the 2nd-Z-15% cargoes, see Supplementary file 5A. DOI: http://dx.doi.org/10.7554/eLife.21184.012

Figure 5.. GO term (Biological Process) enrichments of the 3rd-Z-4% cargoes.

The 3rd-Z-4% cargoes were analyzed for GO term (term type, Biological Process) enrichment. The significantly enriched terms (p<0.05, cyan) in the 3rd-Z-4% cargoes of four or fewer NTRs were selected, and a representative term for each group of highly similar is presented with their p-values and the numbers (#) of cargoes annotated with them. Total No. denotes the number of proteins annotated with each term in the database. Related terms are bundled in the same color. This table was extracted from Supplementary file 6B. All the GO terms annotated to the 3rd-Z-4% cargoes are listed in Supplementary file 7. The correspondence between each 3rd-Z-4% cargo and GO term is summarized in Supplementary file 9. For the 2nd-Z-15% cargoes, see Supplementary files 6A, 8, and 10. DOI: http://dx.doi.org/10.7554/eLife.21184.013

Figure 6.. GO term (Cellular Component and Molecular Function) enrichments of the 3rd-Z-4% cargoes.

The 3rd-Z-4% cargoes were analyzed, and the results are presented in a format similar to that of Figure 5. (A) Term type, Cellular Component. (B) Term type, Molecular Function. These tables were extracted from Supplementary file 6B. All the GO terms annotated to the 3rd-Z-4% cargoes are listed in Supplementary file 7. The correspondence between each 3rd-Z-4% cargo and GO term is summarized in Supplementary file 9. For the 2nd-Z-15% cargoes, see Supplementary file 6A, 8, and 10. DOI: http://dx.doi.org/10.7554/eLife.21184.014

Figure 7.. mRNA processing factors in the 2nd-Z-rankings.

The ranks of the mRNA processing factors in the 2nd-Z-rankings of the 12 NTRs are presented. The color scale is set by percentile rank as indicated. The 2nd-Z-rankings of the 12 NTRs include 275 proteins in total that are annotated with mRNA processing in GO. Of these, 69 were selected and are presented. For other factors and the 3rd-Z-rankings, see Supplementary file 11A. DOI: http://dx.doi.org/10.7554/eLife.21184.015

Figure 8.. Ribosomal proteins in the 2nd-Z-rankings.

The ranks of the ribosomal proteins in the 2nd-Z-rankings of the 12 NTRs are presented. The color scale is set by the percentile rank as indicated. For the 3rd-Z-rankings, see Supplementary file 11B. DOI: http://dx.doi.org/10.7554/eLife.21184.016

Author response image 1.. NTRs in the nuclei of permeabilized cells (Western).

DOI: http://dx.doi.org/10.7554/eLife.21184.029

Author response image 2.. Imp-α1 in depleted cytosolic extract (Western).

DOI: http://dx.doi.org/10.7554/eLife.21184.030

Author response image 3.. NTR concentrations in total and depleted cytosolic extracts (Western).

(Left panel) Reproduced from Kimura et al., 2013, Mol. Cell. Proteomics 12:145-157. DOI: http://dx.doi.org/10.7554/eLife.21184.031