Expanding the Interactome of TES by Exploiting TES Modules with

Subscriber access provided by UNIV OF ARIZONA

Article

Expanding the interactome of TES by exploiting TES modules with different subcellular localizations Stefano Sala, Marleen Van Troys, Sandrine Medves, Marie Catillon, Evy Timmerman, An Staes, Elisabeth Schaffner-Reckinger, Kris Gevaert, and Christophe Ampe J. Proteome Res., Just Accepted Manuscript • DOI: 10.1021/acs.jproteome.7b00034 • Publication Date (Web): 05 Apr 2017 Downloaded from http://pubs.acs.org on April 6, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Journal of Proteome Research is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 53

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Proteome Research

Expanding the interactome of TES by exploiting TES modules with different subcellular localizations Stefano Sala1, Marleen Van Troys1, Sandrine Medves2,4, Marie Catillon2, Evy Timmerman1,3, An Staes1,3, Elisabeth Schaffner-Reckinger2, Kris Gevaert1,3, Christophe Ampe1,* 1

Department of Biochemistry, Ghent University, Belgium

² Cytoskeleton and Cell Plasticity Lab, Life Sciences Research Unit - FSTC, University of Luxembourg, Luxembourg 3

VIB Medical Biotechnology Center, Ghent, Belgium

4

Laboratory of Experimental Cancer Research, LIH, Luxembourg, Luxembourg

(S) supporting information

Abstract: The multimodular nature of many eukaryotic proteins underlies their temporal or spatial engagement in a range of protein co-complexes. Using the multimodule protein testin (TES), we here report a proteomics approach to increase insight in co-complex diversity. The LIM-domain containing and tumor suppressor protein TES is present at different actin cytoskeleton adhesion structures in cells and influences cell migration, adhesion and spreading. TES module accessibility has been proposed to vary due to conformational switching and variants of TES lacking specific domains target to different subcellular locations. By applying iMixPro AP-MS (‘intelligent Mixing of Proteomes’-affinity purification-mass spectrometry) to a set of tagged-TES modular variants, we identified proteins residing in module-specific co-complexes. The obtained distinct module-specific interactomes combine to a global TES interactome that becomes more extensive and richer in information. Applying pathway analysis to the module interactomes revealed expected actin-related canonical pathways and also less expected pathways.

1 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

We validated two new TES co-complex partners: TGFB1I1 and a short form of the glucocorticoid receptor. TES and TGFB1I1 are shown to oppositely affect cell spreading providing biological validity for their co-presence in complexes since they act in similar processes.

Keywords: actin, affinity purification-mass spectrometry, cell spreading, focal adhesion, glucocorticoid receptor, pathway analysis, protein-protein interaction, transforming growth factor beta 1 induced 1, VASP, Zyxin

Introduction The actin cytoskeleton is a complex network of actin polymers and associated proteins. This system relies on dynamic and thus temporal formation of protein complexes which give rise to different subcellular structures such as stress fibres, lamellipodia, filopodia and focal adhesions. Proteins that build up these structures are key players in cellular processes such as migration, invasion, adhesion, spreading and proliferation. Their dysfunction or altered expression is frequently associated with disease, including cancer progression and metastasis1. It is therefore important to define the interactome of these, often multimodular, proteins, preferably in a subcellular spatial manner. Cellular localization of particular actin cytoskeletal proteins is often dictated by so-called closed and open conformations2–4. Different surfaces of the proteins are expected to be exposed after conformational switching, enabling differential interactions with partner proteins. TES is downregulated in human tumours and tumour cell lines and preclinical studies in animals support its role as tumour suppressor 2,5–10. In cells, TES is present at different locations: it resides in the nucleus, in the cytoplasm and in stress fibres, and it is recruited to focal adhesions, both in a zyxindependent as well as a zyxin-independent manner8,9,11,12. More recently, TES has also been observed together with VASP and zyxin at adherens junctions13. Next to zyxin and VASP, TES is known to form complexes with a variety of other actin cytoskeleton proteins (such as Mena, EVL, actin, talin, paxillin, 2 ACS Paragon Plus Environment

Page 2 of 53

Page 3 of 53

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


spectrin and α-actinin2,8,14–16 ) and the interaction repository BioGRID (v 3.4, https://thebiogrid.org/) currently lists 47 interactors based on both reductionist approaches and high throughput studies. As such it can be expected that TES is part of different multicomponent complexes, however TES itself was as yet not used as bait in a proteomics based approach. TES is a modular protein consisting of an N-terminal region rich in cysteine residues, a PET (Prickle, Espinas, TES) domain and three C-terminal zinc finger-containing LIM (Linl-1, Isl-1, Mec-3) domains17,18 (Fig. 1A). For TES, the proposal of an open and closed conformation arises from the observation that the N- and C-terminal parts interact with each other2 and, when expressed separately in cells, these two halves display partly different subcellular localizations with a dominant role for the TES LIM domains in targeting to focal adhesions. This suggests that TES domains within each half have a differential accessibility that depends on the conformation TES adopts, and that may dictate the presence of TES in (partly) different complexes. Here we exploited the modular nature of TES to document its differential co-complexing in relation to its subcellular segregation. Co-complex methods that allow unbiased detection of protein partners classically employ an affinity purification-mass spectrometry (AP-MS) approach: an epitope-tagged bait is enriched from lysates of cells under conditions that preserve the complexes in which the bait resides19. Without proper controls, AP-MS has no strong discriminatory power between specific partners and non-specifically co-purifying proteins20, so that many of the latter are present in final result lists. The latter can be filtered out using dedicated algorithms such as SFINX21. It was, however, recently shown that different algorithms filter out different proteins making in silico scoring for specificity still challenging22. We here opted to use an experimental design that allows a more straightforward selection of specific partners in the obtained protein hit lists. This recently published method, iMixPro23, uses mixing of SILAC labelled proteomes of control and bait-expressing cells to identify peptides from specific binding partners and from proximal proteins, based on a specific light/heavy isotope ratio. This method was developed on HEK cells



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

transiently overexpressing bait proteins and on engineered cell lines expressing epitope-tagged endogenous proteins23. We used the iMixPro AP-MS strategy to analyse GFP-coupled TES-modular variants that show a different subcellular localization in HeLa cells. The presented data show that these modular variants yield differential interactomes containing both known and new components of TES complexes that together form an expansion of the TES interactome. We coupled this to pathway analysis and validated the interaction of two newly identified co-complex proteins: transforming growth factor beta-1-induced 1 (TGFB1I1, ARA55, HIC-5) and a low abundant isoform of the glucocorticoid receptor (GR).

EXPERIMENTAL PROCEDURES TES, GR and TGFB1I1 encoding expression vectors The mammalian expression vectors used contain cDNA that encodes the TES modular variants (with different combinations of the TES domains) coupled to eGFP (see Fig. 1). The cDNAs FL (encoding the full length protein), NT (encoding the CR and PET domains), LIM1-3 (encoding the 3 LIM domains) or PET (encoding the PET-domain) were cloned using the XhoI and BamHI restriction sites in the multiple cloning site of pEGFP-N3 or pEGFP-C3 (Clontech). For the construction of pEGFP-N3-∆PET (encoding a protein variant that lacks the PET domain), the cDNA encoding amino acids 1-91 of TES was inserted using the XhoI and BamHI restriction sites in the multiple cloning site of pEGFP-N3, followed by a second insertion of the cDNA encoding amino acids 200-421 of TES using the BamHI restriction site. For expression of pDsRed-tagged TES variants, the cDNA encoding GFP was replaced by cDNA encoding DsRed in the pEGFP-N3 plasmids and the following plasmids were obtained: pDsRed-FL, pDsRed-NT, pDsRed-LIM1-3. The prokaryotic expression vector pGEX-2TK-∆PET (encoding a protein variant that lacks the PET domain and that is coupled to glutathione-S-transferase GST) was obtained by cloning the cDNAs encoding amino acids 1-91 and amino acids 200-421 of TES (using the BamHI and the BamHI 4 ACS Paragon Plus Environment

Page 4 of 53

Page 5 of 53

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


and SmaI restriction sites, respectively) in the multiple cloning site of pGEX-2TK (GE Healthcare). All constructs were verified by sequencing. pEGFP-C1-TGFB1I1 (encoding mouse TGFB1I1 coupled to GFP) was kindly provided by Christopher Turner of the SUNY Upstate Medical University of New York. The mouse TGFB1I1 cDNA is present between the BamHI and XbaI restriction sites in the multiple cloning site. pEF-FLAG-hGRα and pEF-FLAG-hGRβ vectors were kindly provided by Karolien De Bosscher of the Cytokine Receptor Lab at Ghent University (Belgium); human GRα or GRβ cDNA is present between the NcoI and XbaI restriction sites. Cell culture and transient transfection HeLa cells and HEK293T cells were cultured in Dulbecco’s Modified Eagle Medium with GlutaMax, supplemented with 10% fetal bovine serum and 1% penicillin/streptomycin (Gibco, Life technologies) at 37°C and 5% CO2. For transfection, a calcium phosphate-mediated transfection was used. For immunofluorescence, cells in a 6-well plate (1.2x105 cells/well) on coverslips were transfected with 5 µg of one of the above described plasmids. HeLa cells used for Co-IP were transfected at 70-80% confluency in a 175 cm² falcon with 50 µg of one of the pEGFP-TES encoding plasmids. Co-immunoprecipitation Cells were lysed for 20 min on ice in lysis buffer containing 50 mM Tris HCl (pH 7.6), 125 mM NaCl, 5% glycerol, 0.2 % NP40 (Sigma), 1.5 mM MgCl2, phosphatase inhibitor cocktail III (Calbiochem) and complete protease inhibitor cocktail (Roche). The lysates were centrifuged at 12000 g for 15 min. Cleared lysate containing 1 mg of total protein was incubated with 25 µl GFP Trap®_beads (anti-GFP nanobody coupled agarose beads from Chromotek) for 2 h at 4°C. Beads were washed three times in NP40 based lysis buffer and heated for 5 min at 95°C in SDS containing sample buffer (65 mM Tris HCl (pH 6.8), 5% SDS, 250 mM dithiothreitol, 20% glycerol and 0.2% bromophenol blue), or processed for proteomics analysis (see below).



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

For reverse Co-IP: lysates (containing 1 mg of total protein) of cells expressing a TES-GFP protein variant were incubated for 2 h at 4°C with 5 µg of anti-GR (H300, Santa Cruz), anti-TGFB1I1 (H75, Santa Cruz), anti-GRα (Abcam, ab3580) or anti-TES (Abcam, ab57292) antibody. Subsequently, lysates were incubated for 2 h at 4°C with 20 µl G-protein sepharoseTM 4 fast flow beads (GE Healthcare). Beads were washed 3 times in NP40-based lysis buffer and heated for 5 min at 95°C in SDS containing sample buffer. Proteomics: sample preparation, data analysis, statistics and pathway analysis To identify specific interaction partners of the TES modular variants we used the iMixPro method as described in23 with a few adaptations. HeLa cells were cultured in medium containing SILAC DMEM (Silantes) with 10% dialysed fetal bovine serum, 0,5% penicillin/streptomycin, 1% glutamax (Gibco, Life technologies) and supplemented with either light (12C6) arginine (final concentration 84 g/l) and lysine (73g/l) or heavy (13C6) arginine (84 g/l) and lysine (73g/l) (see below). After 10 days of cell expansion in light or heavy SILAC medium, labelled HeLa cells were transfected as described above with plasmids encoding TES variants coupled to a GFP tag (Fig.1A). According to the iMixPro concept, each experiment consists of five conditions as shown schematically in Suppl. Fig. 1: TES-GFP variant expressing cells once in light medium (containing 12C6 arginine and lysine) and once in heavy medium (containing 13C6 arginine and lysine) and control (GFP) expressing cells three times in light medium. 48 h after transfection, cells were lysed with NP40 based lysis buffer (see above) and five separate Co-IPs were performed (Suppl. Fig. 1): lysate containing 1 mg of protein of each of the five conditions was incubated with 25 µl GFP Trap®_A beads (Chromotek) and incubated for 3 h at 4°C. After incubation, beads were transferred to a Poly-Prep® chromatography Column (Biorad), washed with 100 mM NaCl, 50 mM ammonium bicarbonate (Sigma). Elution was performed in 4% formic acid. At this point, the five samples (4 light conditions and 1 heavy condition) were mixed and immediately processed for tryptic digest23. Samples were acidified with 10% TFA (trifluoroacetic acid) and analysed by LC-MS/MS analysis on an Ultimate 3000 RSLC nano LC (Thermo Fisher Scientific, Bremen, Germany) in-line connected to a Q 6 ACS Paragon Plus Environment

Page 6 of 53

Page 7 of 53

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Exactive mass spectrometer (Thermo Fisher Scientific) or a LTQ-Orbitrap XL (Thermo Fisher Scientific). The peptides were first loaded on a trapping column (made in-house, 100 µm internal diameter (I.D.) × 20 mm, 5 µm beads C18 Reprosil-HD, Dr. Maisch, Ammerbuch-Entringen, Germany) and after flushing from the trapping column, peptides were loaded on an analytical column (made in-house, 75 µm I.D. × 150 mm, 5 µm beads C18 Reprosil-HD, Dr. Maisch) packed in the needle. Peptides were loaded with loading solvent (0.1% trifluoroacetic acid in water/ acetonitrile, 2/98 (v/v)) and separated with a linear gradient from 98% solvent A (0.1% formic acid in water) to 55% solvent B (0.1% formic acid in water/acetonitrile, 20/80 (v/v)) in 30 min at a flow rate of 300 nL/min, followed by a 5 min wash reaching 99% solvent B. The mass spectrometer was operated in data-dependent, positive ionization mode, automatically switching between MS and MS/MS acquisition for the 10 most abundant peaks in a given MS spectrum. The source voltage was 3.4 kV, and the capillary temperature was 275°C. One MS1 scan (m/z 400−2,000, AGC target 3 × 106 ions, maximum ion injection time 80 ms), acquired at a resolution of 70,000 (at 200 m/z), was followed by up to 10 tandem MS scans (resolution 17,500 at 200 m/z) of the most intense ions fulfilling predefined selection criteria (AGC target 5 × 104 ions, maximum ion injection time 60 ms, isolation window 2 Da, fixed first mass 140 m/z, spectrum data type: centroid, underfill ratio 2%, intensity threshold 1.7xE4, exclusion of unassigned, 1, 5-8, >8 positively charged precursors, peptide match preferred, exclude isotopes on, dynamic exclusion time 50 s). The HCD collision energy was set to 25% normalized collision energy and the polydimethylcyclosiloxane background ion at 445.120025 Da was used for internal calibration (lock mass). The LTQ-Orbitrap XL mass spectrometer was operated in datadependent, position ionization mode, automatically switching between MS and MS/MS acquisition for the five most abundant peaks in a given MS spectrum. In the LTQ-Orbitrap XL, full scan MS spectra were acquired in the Orbitrap at a target value of 1E6 with a resolution of 60000. The five most intense ions were then isolated for fragmentation in the linear ion trap, with a dynamic exclusion of 60 s. Peptides were fragmented after filling the ion trap at a target value of 1E4 ion counts23.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

From the MS/MS data in each LC run, Mascot Generic Files were created using the Mascot Distiller software (version 2.5.1, Matrix Science). These peak lists were searched using the Mascot search engine with the Mascot Daemon interface (version 2.5.1, Matrix Science). Spectra were searched against the Swiss-Prot database (release 2015_11) with a taxonomy filter on Homo sapiens (concatenated with the sequence of GFP) containing 20100 entries. Variable modifications were set to pyro-glutamate formation of amino-terminal glutamine, acetylation of the protein N-terminus and methionine oxidation to methionine-sulfoxide. Labelling was configured allowing for light or heavy labeled arginine (13C6) or lysine (13C6). This setting was configured in exclusive mode and set as quantitation method in the Mascot search parameters. Mass tolerance on precursor ions was set to ± 10 ppm (with Mascot’s C13 option set to 1), and on fragment ions to 20 mmu. The instrument setting was on ESI-QUAD. The enzyme was set to trypsin/P, allowing for 1 missed cleavage. Only peptides that were ranked first and scored above the threshold score, set at 99% confidence, were withheld using the ms_lims software package24. All PSMs (peptide spectrum matches) matching peptides shorter than 8 amino acids were rejected. The identified peptides were then quantified using Mascot Distiller Toolbox version 2.5.1 (MatrixScience) in the precursor protocol and the Trapezium rule was used as integration method. As filters for the quantitation a charge state threshold of 2, a matched Rho of 0.97, an XIC threshold of 0.3, an XIC maximum width of 250, an XIC smoothing of 1 and an elution profile correlation threshold of 0.16 were used. The quantified peptides of each iMixPro experiment generated a list of proteins that were present after Co-IP and their corresponding light/heavy (L/H) ratios (for protein lists see Suppl. data S1). This list contains also proteins identified by only one peptide (to have sufficient entries for subsequent IPA pathway analysis). The lists were uploaded to PRIDE via Proteomexchange and carry PXD005058 as dataset identifier (http://www.ebi.ac.uk/pride). All data management was done by ms_lims24, and data analysis was performed using R (http://www.R-project.org) embedded in KNIME25. We calculated the medians and standard deviations of


Page 8 of 53

Page 9 of 53

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


the mixed normal distributions of the log2 of the ratios with the “normalmixEM” function embedded in the “mixtools” R package. However, this approach did not allow deciding to which distribution a L/H ratio belongs because histograms of identified proteins versus the log2 of their L/H ratios (not shown) did not show the theoretically expected separation in two sub-distributions. This is likely due to differences in the expression levels of the bait protein in the differently labelled cell lines (Suppl. Fig. 1, Suppl. data S2). We therefore used an alternative statistical approach on the total distributions to extract lists of specifically associated proteins which is detailed in Suppl. data S2. In short, specifically associated proteins were extracted based on having a Log2 L/H ratio

Expanding the Interactome of TES by Exploiting TES Modules with

Recommend Documents