Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
Filters applied. Clear all
, 8, 15690

Split-BioID a Conditional Proteomics Approach to Monitor the Composition of Spatiotemporally Defined Protein Complexes


Split-BioID a Conditional Proteomics Approach to Monitor the Composition of Spatiotemporally Defined Protein Complexes

Isabel Myriam Schopp et al. Nat Commun.


Understanding the function of the thousands of cellular proteins is a central question in molecular cell biology. As proteins are typically part of multiple dynamic and often overlapping macromolecular complexes exerting distinct functions, the identification of protein-protein interactions (PPI) and their assignment to specific complexes is a crucial but challenging task. We present a protein fragments complementation assay integrated with the proximity-dependent biotinylation technique BioID. Activated on the interaction of two proteins, split-BioID is a conditional proteomics approach that allows in a single and simple assay to both experimentally validate binary PPI and to unbiasedly identify additional interacting factors. Applying our method to the miRNA-mediated silencing pathway, we can probe the proteomes of two distinct functional complexes containing the Ago2 protein and uncover the protein GIGYF2 as a regulator of miRNA-mediated translation repression. Hence, we provide a novel tool to study dynamic spatiotemporally defined protein complexes in their native cellular environment.

Conflict of interest statement

The authors declare no competing financial interests.


Figure 1
Figure 1. Design of split-BioID.
(a) Schematic drawing of the proof-of-principle set-up for split-BioID. FRB fused to CBirA* and FKBP fused to NBirA* interact with each other on addition of rapamycin. (b) BirA crystal structure (Protein Data Bank 1BIB) showing the four tested splitting sites (1: Q65/L66, 2: E256/G257, 3: N270/F271 and 4: N273/R274), the optimal E256/G257 site is indicated (See Supplementary Fig.1).
Figure 2
Figure 2. Characterization of split-BioID.
(a) The four tested constructs, corresponding to all possible orientations of the CBirA*-FRB and NBirA*-FKBP fusions. (b) Blots of lysates of HeLa cells transiently transfected with the constructs shown in a, and treated with or without rapamycin. Biotinylation was analysed using Alexa680-labelled streptavidin, and expression levels of the fusion proteins with antibodies against FLAG and Myc tags as indicated. (c) Quantification of b: relative overall biotinylation levels were estimated by integrating Alexa680 signals in each lane in b and normalizing them to the signal measured for an endogenous biotinylated protein in the same lane. Signals from non-transfected samples were set to 0% (error bars, s.d.; n=3 independent experiments). (d) Activity of the original BirA* compared to the split-BirA* (construct 1). Total biotinylation (streptavidin-Alexa680) and protein expression levels (anti-Myc signal) were set to 100% for BirA*, taking into account the different protein amounts loaded. Relative activity is the ratio of biotinylation over Myc levels.
Figure 3
Figure 3. Application of split-BioID to a physiological phosphorylation-dependent PPI.
(a) Blots of lysates of cells transiently expressing NBirA*-14-3-3ɛ/CBirA*-Cdc25c (WT or S216A), or NBirA*-GFP/CBirA*-Cdc25c WT as a control. (b) Quantitative analysis of a performed as described in Fig. 2c (error bars, s.d.; n=3 independent experiments). (c) Volcano plot showing proteins enriched in the Cdc25C WT over the control BioID samples from stable cell lines. The logarithmic ratios of protein LFQs were plotted against negative logarithmic P values of a two-sided two samples t-test. The hyperbolic curve delimitate significantly enriched proteins from common hits (FDR≤0.07, n=3). Hits that showed higher LFQs than in the GFP and the Cdc25C S216A samples are indicated in red. (d) Validation co-IP experiments from HeLa cell lysates transfected with myc-NBirA*-14-3-3ɛ and FLAG-CBirA*-Cdc25C or lysates of untransfected cells as a control. Endogenous LMO7 is detected in both FLAG and Myc IPs from transfected cells while endogenous CKAP5 is not.
Figure 4
Figure 4. Localization of the fusion proteins used for the Cdc25C/14-3-3ɛ split-BioID.
Immunofluorescence of transiently expressed constructs by Myc- and FLAG-tag detection as well as the detection of biotinylated proteins by Cy5-coupled streptavidin. Scale bar, 15 μm.
Figure 5
Figure 5. Monitoring maturation of the miRISC with split-BioID.
(a) Schematic of the miRNA-mediated silencing pathway. Main protein components of the RLC and miRISC are indicated. Dark grey circles are the proteins used for split-BioID. (b) Left, blots of lysates of cells transfected with the indicated combinations of fusion proteins. CBirA*-Ago2 combined with NBirA* fused to Dicer, TNRC6C or GFP showed the best interaction-induced biotinylation and was thus used further (See Supplementary Fig. 4). Right, immunoprecipitation of Myc-tagged NBirA*-TNRC6C from transfected or control cells. (c) Streptavidin capture of the biotinylated proteins from the CBirA*-Ago2/NBirA*-Dicer, -TNRC6C and -GFP samples. As controls, NBirA*-Dicer and -TNRC6C were also expressed in the absence of CBirA* fragments (replaced by FRB). CNOT1 and GIGYF2 were specifically detected in the Ago2/TNRC6C sample and TRBP in the Ago2/Dicer eluate.
Figure 6
Figure 6. Localization of the fusion proteins used for the Cdc25C/14-3-3ɛ split-BioID.
Immunofluorescence of transiently expressed constructs by Myc- and FLAG-tag detection as well as the detection of biotinylated proteins by Cy5-coupled streptavidin. Scale bar, 15 μm. All constructs are expressed in the cytosol and have the chance to interact with each other.
Figure 7
Figure 7. Main hits enriched in Ago2/Dicer and Ago2/TNRC6C split-BioID.
Volcano plot showing proteins enriched in the TNRC6C (a) or Dicer (b) sample over the control BioID samples as in Fig. 3c. Hits that showed higher LFQs than in the GFP and the Dicer samples (a) or in the GFP and TNRC6C samples (b) are indicated in red. For clarity, not all hits are labelled (c) Selected hits from the MS analysis of Ago2/Dicer (right) anf Ago2/TNRC6C (left) split-BioID samples. The complete list is given in Supplementary Data 2.
Figure 8
Figure 8. Identification of a novel miRISC factor.
(a) Immunofluorescence of endogenous Ago2 and GIGYF2 in HeLa 11ht cells. Scale bar, 15 μm. (b) Blots of HeLa cell lysates immunoprecipitated with the indicated antibodies. The blots were decorated with antibodies directed against Ago2, TNRC6A or GIGYF2 as indicated. When indicated (+/− MNase), lysate were pre-treated with micrococcal nuclease. This panel is representative of three different experiments. (c) Schematic representation of TNRC6C and GIGYF2 and fragments used for the in vitro binding assay. Main domains are highlighted as well as the PPGL motif within TNRC6C. N-GW: N-terminal GW-rich region, UBA: ubiquitin associated-like domain, PAM2: poly(A) binding protein (PABP)-interacting motif 2, RRM: RNA-recognition motif, CED: C-terminal effector motif. (d) His6 pulldowns on Ni-NTA beads using recombinant MBP-CED-His6 (WT or AAGL variant) and GST-tagged fragment of GIGYF2 comprising the GYF domain (532–740) or not (607–740).
Figure 9
Figure 9. Validation of novel miRISC-regulating factor.
(a) Dual luciferase assay following knockdown experiments with the indicated siRNA pools performed in stable inducible cell lines expressing the depicted renilla luciferase reporters. The bars show the expression levels of the RL-WT reporter (with miRNA-binding sites) normalized to a co-expressed control firefly luciferase reporter at the indicated time points after induction. Expression levels are indicated as percentages of values for the matching non-repressed MUT reporter (with no miRNA-binding sites), which are set to 100% at each time point. At 2 h post induction, miRNA-mediated repression is significantly alleviated on knockdown of GIGYF2. P value was calculated using a two-tailed student's paired t-test (n=6, error bars are s.e.m.). (b) Western blots showing knockdown efficiency of the siRNAs used in a. Unspecific bands are labelled with a star.

Similar articles

See all similar articles

Cited by 27 PubMed Central articles

See all "Cited by" articles


    1. Jonas S. & Izaurralde E. Towards a molecular understanding of microRNA-mediated gene silencing. Nat. Rev. Genet. 16, 421–433 (2015). - PubMed
    1. Frohn A. et al. . Dicer-dependent and -independent Argonaute2 protein interaction networks in mammalian cells. Mol. Cell Proteomics 11, 1442–1456 (2012). - PMC - PubMed
    1. Landthaler M. et al. . Molecular characterization of human Argonaute-containing ribonucleoprotein complexes and their bound target mRNAs. RNA 14, 2580–2596 (2008). - PMC - PubMed
    1. Li S. et al. . TRIM65 regulates microRNA activity by ubiquitination of TNRC6. Proc. Natl Acad. Sci. USA 111, 6970–6975 (2014). - PMC - PubMed
    1. Braun J. E., Huntzinger E., Fauser M. & Izaurralde E. GW182 proteins directly recruit cytoplasmic deadenylase complexes to miRNA targets. Mol. Cell 44, 120–133 (2011). - PubMed

Publication types

LinkOut - more resources