Mapping the MOB proteins' proximity network reveals a unique interaction between human MOB3C and the RNase P complex

Abstract

Distinct functions mediated by members of the monopolar spindle-one-binder (MOB) family of proteins remain elusive beyond the evolutionarily conserved and well-established roles of MOB1 (MOB1A/B) in regulating tissue homeostasis within the Hippo pathway. Since MOB proteins are adaptors, understanding how they engage in protein-protein interactions and help assemble complexes is essential to define the full scope of their biological functions. To address this, we undertook a proximity-dependent biotin identification approach to define the interactomes of all seven human MOB proteins in HeLa and human embryonic kidney 293 cell lines. We uncovered >200 interactions, of which at least 70% are unreported on BioGrid. The generated dataset reliably recalled the bona fide interactors of the well-studied MOBs. We further defined the common and differential interactome between different MOBs on a subfamily and an individual level. We discovered a unique association between MOB3C and 7 of 10 protein subunits of the RNase P complex, an endonuclease that catalyzes tRNA 5' maturation. As a proof of principle for the robustness of the generated dataset, we validated the specific interaction of MOB3C with catalytically active RNase P by using affinity purification-mass spectrometry and pre-tRNA cleavage assays of MOB3C pulldowns. In summary, our data provide novel insights into the biology of MOB proteins and reveal the first interactors of MOB3C, components of the RNase P complex, and hence an exciting nexus with RNA biology.

Keywords: BioID; MOB proteins; MOB2; MOB3C; MOB4; RNase P complex; proteomics; proximity labeling techniques.

Figure 1

BioID proximity labeling screens for the MOB proteins.A, schematic outline of the BioID screens’ pipeline. B, Western blot analysis of the parental and BioID bait-expressing Flp-In T-REx HeLa and HEK 293T cells. C, confocal microscopy images of the bait’s expression (anti-FLAG) and biotinylation (streptavidin) patterns for the seven MOB proteins and a control condition (cells expressing BirA-FLAG-EGFP) in HeLa cells. B and C, cells were treated with tetracycline (to induce expression) and biotin (to induce biotinylation) for 24 h. BioID, biotinylation identification; EGFP, enhanced GFP; HEK, human embryonic kidney cell line; MOB, monopolar spindle-one-binder.

Figure 2

BioID recalls the bona fide MOB protein interactors and extends their proximity network.A, Venn diagrams showing the number of MOB preys identified in the two cell lines used in BioID. B, histogram demonstrating an enumeration of interactors identified for every MOB protein in our BioID screens (BFDR ≤ 0.01) in addition to a representation of the BioGrid recalls. C, dot plots highlighting the bona fide interactors of MOB1A/B and MOB2 in HeLa and HEK293 cells. BFDR, Bayesian false discovery rate; BioID, biotinylation identification; HEK293, human embryonic kidney 293 cell line; MOB, monopolar spindle-one-binder.

Figure 3

Overview of the interactome of the MOB subfamilies.A, heatmap highlighting the relative MoSS score for the bona fide interactors of MOB1A/B, MOB2, and MOB4. B, an UpSet plot enumerating intersections of nonredundant proximal interactions identified for different MOB baits in either HeLa or HEK293 cells by BioID. C, heatmap highlighting the preys shared between MOB3A/B/C and other MOB subfamilies in addition to their individual common preys. HEK293, human embryonic kidney 293 cell line; MOB, monopolar spindle-one-binder; MoSS, MOB specificity score.

Figure 4

BioID reveals novel interactors of the MOB proteins.A–C, heatmaps highlighting proteins in the proximity of MOB1A/B, MOB2 (A), MOB4 (B), and MOB3A/B/C (C). D–G, dot plots displaying selected terms from over-represented CORUM complexes (D), Gene Ontology (GO) cellular components (E), GO biological processes (F), and KEGG pathways (G). BioID, biotinylation identification; KEGG, Kyoto Encyclopedia of Genes and Genomes; MOB, monopolar spindle-one-binder.

Figure 5

MOB3C interacts with catalytically active RNase P complex.A, coimmunoprecipitation assay for YFP-tagged POP1 and RPP30 with FLAG-tagged MOB3C in HeLa cells. B, schematic outline for the affinity purification–mass spectrometry (AP–MS) with DSP crosslinking (CL). A Western blot analysis validating the used cell lines in AP–MS is shown. C, Venn diagram showing the commonly defined proteins in the BioID and crosslinked AP–MS datasets. The table displays the average spectral counts of each RNase P subunit defined in the AP–MS dataset. D, a proximity network of the RNase P protein subunits defined in both the MOB3C BioID and AP–MS datasets in reference to BioGrid (see the Experimental procedures section). E, pre-tRNA cleavage assay of the pull-down experiments using GST-MOB1A or GST-MOB3C. Time-course reactions showing RNase P activity when GST-MOB3C, but not GST-MOB1A, was used. +: a positive control of pre-tRNA^Arg cleaved by recombinant Escheichia coli RNase P; –: a negative control of pre-tRNA^Arg without enzyme; I: input. F, SDS-PAGE analysis confirms the presence of GST-MOB3C and GST-MOB1A post-pulldown. DSP, dithiobis (succinimidyl propionate); EV, empty vector; GST, glutathione-S-transferase; MOB, monopolar spindle-one-binder.