Proximity labeling by a recombinant APEX2-FGF1 fusion protein

Subscriber access provided by UNIV OF NEW ENGLAND ARMIDALE

Article

Proximity labeling by a recombinant APEX2-FGF1 fusion protein reveals interaction of FGF1 with the proteoglycans CD44 and CSPG4. Yan Zhen, Ellen Margrethe Haugsten, Sachin Kumar Singh, and Jørgen Wesche Biochemistry, Just Accepted Manuscript • Publication Date (Web): 29 May 2018 Downloaded from http://pubs.acs.org on May 29, 2018

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 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 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.

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 30 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

Biochemistry

Proximity labeling by a recombinant APEX2-FGF1 fusion protein reveals interaction of FGF1 with the proteoglycans CD44 and CSPG4.

Yan Zhen1,2, Ellen Margrethe Haugsten2,3, Sachin Kumar Singh2,3 and Jørgen Wesche2,3,*

1

Department of Molecular Cell Biology, Institute for Cancer Research, The Norwegian Radium

Hospital, Oslo University Hospital, Montebello, 0379 Oslo, Norway 2

Centre for Cancer Cell Reprogramming, Institute of Clinical Medicine, Faculty of Medicine,

University of Oslo, Montebello, 0379 Oslo, Norway 3

Department of Tumor Biology, Institute for Cancer Research, The Norwegian Radium

Hospital, Oslo University Hospital, Montebello, 0379 Oslo, Norway

* Author to whom correspondence should be addressed:

Jørgen Wesche Tel: +47 22 78 19 31 Fax: +47 22 78 18 45 Email: [email protected]

Running title: Proteomics reveal binding of FGF1 to CD44

1

ACS Paragon Plus Environment

Biochemistry 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

Page 2 of 30

ABSTRACT Fibroblast growth factor 1 (FGF1) binds to specific FGF receptors (FGFRs) at the surface of target cells to initiate intracellular signaling. While heparan sulfate proteoglycans (HSPG) are well described co-receptors, it is uncertain if there are additional binding sites for FGF1 at the cell surface. To address this, we have devised and tested a method to identify novel binding sites for FGF1 at the cell surface, which may also be applicable for other protein ligands. We constructed an APEX2-FGF1 fusion protein to perform proximal biotin labeling of proteins following binding of the fusion protein to the cell surface. After functional validation of the fusion protein by a signaling assay, we used this method to identify binding sites for FGF1 on cell surfaces of living cells. We confirmed the feasibility of our approach by detection of FGFR4, a well-known and specific receptor for FGF1. We subsequently screened for novel interactors using RPE1 cells and identified the proteoglycans CSPG4 (NG2) and CD44. We found that FGF1 bound CD44 through its heparin-binding moiety. Moreover, we found that FGF1 co-localized with both CSPG4 and CD44 at the cell surface suggesting that these receptors act as storage molecules that create a reservoir of FGF1. Importantly, our data demonstrate that recombinant ligand-APEX2 fusion proteins can be used to identify novel receptor interactions on the cell-surface.

2


Page 3 of 30 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

Biochemistry

INTRODUCTION Fibroblast growth factor 1 (FGF1) binds to FGF receptors (FGFRs) on the cell surface of target cells. There are four FGFRs (FGFR1-4) and FGF1 binds to all of them, while other FGF ligands (FGF2-22) bind with varying affinities to the four receptors1. Binding of FGF1 to the extracellular part of FGFRs causes receptor dimerization resulting in auto-transphosphorylation of tyrosines by the intracellular kinase domain2. This initiates phospho-tyrosine based downstream signaling cascades such as the MAPK, AKT, and PLCγ pathways3, 4. In addition to FGFRs, FGF1 binds to heparan sulfate proteoglycans (HSPGs)5, 6. These consist typically of a core protein and glycosaminoglycan (GAG) chains that are attached to the extracellular domain of the protein. The GAGs are sulfated creating a negatively charged molecule that can bind to many positively charged proteins; including FGFs. Proteoglycans act as co-receptors that are utilized to store FGF1 in the extracellular matrix. GAGs are also crucial for the interaction of FGFs with the receptors as they stabilize the complex by binding both FGFs and FGFRs7. Finally, the interaction of FGF1 with GAGs results in strongly increased stability of the protein8. The heparin binding domain of FGF1 has been described in three-dimensional crystal structures, revealing a relatively flat primary interaction surface with the possible contribution of additional secondary binding sites7, 9. There are many classes of proteoglycans that have the potential to bind FGF1 as they are negatively charged and expressed at the cell-surface. However, in addition to electrostatic interactions, the fine structure of the glycosaminoglycan chains (e.g. pattern of sulfation) may facilitate binding to FGF110. FGFs may therefore display selectivity towards certain species of proteoglycan 10, 11.

3


Biochemistry 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

Page 4 of 30

The HSPG family consists of syndecans, glypicans and the soluble perlecans6. FGF1 has been proposed to interact with several HSPGs (GPC1, GPC3, GPC6, SDC2 and SDC4)12. FGF2 in addition to HSPGs, has been shown to bind chondroitin sulfate proteoglycans (CSPGs)13. Other proteoglycans may bind FGFs and given the essential role of proteoglycans in FGF biology6, it will be important to map the different species of FGF binding proteoglycans and their different cell type specific expression. Cells from various cancer types have been shown to overexpress distinct proteoglycans. As this could influence growth factor binding and signaling, there is a need to develop methods that can readily identify the interaction of FGFs with this large group of glycoproteins. We have here tested the possibility of using proximal biotin labeling for such interaction studies. The modified soybean peroxidase, APEX2, effectively biotinylates proteins in living cells14. APEX2 uses biotin-phenol and H2O2 which are supplemented to the cell cultures to tag proteins in close proximity with biotin. Biotin tagged proteins are then readily purified by streptavidin-Sepharose and subsequently identified by mass spectrometry (MS). In this study, APEX2 fused to FGF1 (APEX2-FGF1) was expressed as a recombinant fusion protein and supplemented to cells in culture. Addition of recombinant APEX2-FGF1 to cell cultures allowed us to detect binding sites on the surface of living cells. Via this approach we identified and validated a facultative proteoglycan, CD4415, as a new binding site for FGF1 on the cell surface. The method should be useful to test other FGF ligands and also other protein ligands interacting with cell surface receptors.

4


Page 5 of 30 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

Biochemistry

MATERIAL AND METHODS Plasmids, Antibodies and Compounds APEX2-FGF1 was generated by cloning a PCR fragment containing the APEX2 open reading frame, a linker consisting of nucleotides coding for the amino acids GlyGlyAlaGlyGly and NcoI flanking sites into pET21d-FGF1 cut with NcoI. pET21d-FGF1-myc5 was made previously16. pcDNA3 APEX2-NES was used as template and was a gift from Alice Ting (Addgene plasmid #49386). The following antibodies were used: mouse anti-phospho-FGFR (Tyr653/654) (#3476), mouse anti-CD44 (#3570), mouse anti-phospho-ERK1/2 (Thr202/Tyr204) (#9106), and rabbit anti-ERK1/2 (#9102) from Cell signaling; rabbit anti-FGFR4 (C16) (sc-124), goat anti-FGF1 (sc-1884) and normal mouse IgG control (sc-2025) from Santa Cruz; GAPDH-HRP (ab9482) and goat anti-myc (ab9132) from abcam; mouse anti-γ-tubulin (T6557) from Sigma-Aldrich; rabbit anti-myc (9E10 poly, raised against EQKLISEEDL) from Covance; mouse anti-myc (05724) from Millipore; Streptavidin-HRP (016-030-084) and all secondary antibodies from Jackson Immuno-Research Laboratories. Mouse anti-NG2 (CSPG4) (9.2.27AB) was a gift from Prof. Øystein Fodstad and Prof. Gunhild Mælandsmo from the Department of Tumor biology, the Norwegian Radium Hospital. Protease inhibitor cocktail tablets (ethylenediaminetetraacetic acid (EDTA)-free, complete) were from Roche Diagnostics. Hoechst 33342 and Dynabeads® Protein G was purchased from Life Technologies. Streptavidin Sepharose High Performance was from GE Healthcare Life Sciences. Mowiol, heparin (H3393), and phosphatase inhibitors, Hydrogen peroxide solution, Sodium ascorbate, Trolox were from Sigma-Aldrich. Biotin-Phenol was from Iris Biotech (LS-3500). Bafilomycin A1 was from Calbiochem.

5


Biochemistry 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

Page 6 of 30

Protein expression APEX2-FGF1, FGF1-5Xmyc and FGF1 proteins were prepared as previously described16. Briefly, proteins were expressed in E.coli BL21 (DE3) pLysS with a yield of 20–40 mg of pure protein/liter of culture. A heparin-Sepharose column on an Äkta system (GE Healthcare) was used to purify the proteins by a NaCl gradient. Purity and size of the purified proteins were confirmed by SDS-PAGE.

siRNAs ON-TARGETplus human CD44 siRNA SMARTpool (L-009999-00-0005) and ONTARGETplus non-targeting siRNA (D-001810-01-20) was purchased from Dharmacon (L009999-00-0005).

Cells and Transfection RPE1 and U2OS cells were propagated in DMEM12 or DMEM (respectively) supplemented with 10% (v/v) fetal bovine serum, 100 U/ml penicillin, and 100 µg/ml streptomycin in a 5% (v/v) CO2 atmosphere at 37°C. U2OS cells stably expressing FGFR4 (U2OS-FGFR4) have been described previously17. siRNA transfection was performed using Lipofectamine RNAiMAX transfection reagent (Invitrogen, Life Technologies) according to the manufacturer’s protocol. 20 nM of siRNA was used to deplete CD44, and the experiments were performed 72 hours after transfection.

In vitro peroxidase activity

6


Page 7 of 30 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

Biochemistry

To test the peroxidase activity of APEX2-FGF1, 1 µl 20 ng/µl APEX2-FGF1 was mixed with 20 µl SuperSignal West Dura Extended Duration Chemiluminescent Substrate (Thermo Scientific) and spotted on a filter paper. Chemiluminscence was then detected by Gel Doc (BioRad).

Affinity capture of biotinylated proteins for mass spectrometry 500

µM

biotin-phenol

was

added

to

RPE/U2OS-FGFR4

cells

cultured

in

DMEM12/DMED complete medium at 37°C for 1 hour. Then cells were treated with FGF1 100 ng/ml (7 µM) or 300 ng/ml APEX2-FGF1 (7 µM) in the absence or presence of 20 U/ml heparin for indicated periods of time according to the experimental design. H2O2 1 mM was supplied for one minute to initiate biotinylation. The cells were then washed three times in Quench buffer (10 mM sodium azide, 10 mM sodium ascorbate, 5 mM Trolox in PBS) to terminate the reaction. Next, the cells were scraped in Quenchbuffer and collected for centrifugation at 4000 rpm for 5 minutes. The cell pellets were lysed in Quench lysis buffer (50 mM Tris, pH 7.4, 150 mM NaCl, 0.1% (w/v) SDS, 0.5% (w/v) sodium deoxycholate, 1% (v/v) Triton X-100, 1 mM PMSF (phenylmethylsulfonyl fluoride), 10 mM sodium azide, 10 mM sodium ascorbate, 5 mM Trolox supplemented with protease inhibitors). After 20 minutes on ice, lysates were centrifuged at 14 000 rpm for 10 minutes at 4°C. Supernatants were desalted with Zeba Spin Desalting columns before incubation with 300 µl Streptavidin Sepharose High Performance for 2 hours at 4°C. The beads were washed twice for 5 minutes at 4°C in 1% (v/v) Triton X-100, 500 mM NaCl, 1 mM EDTA, twice for 5 minutes at 4°C in PBS with 0.1% (v/v) Triton X-100, twice for 5 minutes at room temperature in 2% SDS (w/v), twice for 5 minutes at room temperature in 6 M Urea in

7


Biochemistry 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

Page 8 of 30

PBS, six times for 5 minutes at room temperature in 1 M NaCl, 25% (v/v) Acetonitrile and twice for 5 minutes at room temperature in 20% (v/v) Acetonitrile.

Mass Spectrometry Samples were prepared for mass spectrometry as previously described18. For whole cell proteomics, 15 µg protein sample was treated with DTT and IAA and subsequently digested with 1 µg of trypsin according to the FASP protocol 19. All experiments were performed on an Easy nLC1000 nano-LC system connected to a quadrupole - Orbitrap (Q Exactive) mass spectrometer (ThermoElectron, Bremen, Germany) equipped with a nanoelectrospray ion source (EasySpray/Thermo). For liquid chromatography separation we used an EasySpray column (C18, 2 µm beads, 100 Å, 75 µm inner diameter) (Thermo) capillary of 25 cm bed length. The flow rate used was 300 nL/min, and the solvent gradient was 2% (v/v) B to 30% (v/v) B in 120 minutes, then 90% (v/v) B wash in 20 minutes. Solvent A was aqueous 0.1% (v/v) formic acid, whereas solvent B was 100% (v/v) acetonitrile in 0.1% (v/v) formic acid. Column temperature was kept at 60oC. The mass spectrometer was operated in the data-dependent mode to automatically switch between MS and MS/MS acquisition. Survey full scan MS spectra (from m/z 400 to 1,200) were acquired in the Orbitrap with resolution R=70,000 at m/z 200 (after accumulation to a target of 3,000,000 ions in the quadruple). The method used allowed sequential isolation of the most intense multiply-charged ions, up to ten, depending on signal intensity, for fragmentation on the HCD cell using high-energy collision dissociation at a target value of 100,000 charges or maximum acquisition time of 100 ms. MS/MS scans were collected at 17,500 resolution at the Orbitrap cell. Target ions already selected for MS/MS were dynamically excluded for 30

8


Page 9 of 30 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

Biochemistry

seconds. General mass spectrometry conditions were: electrospray voltage, 2.1 kV; no sheath and auxiliary gas flow, heated capillary temperature of 250oC, normalized HCD collision energy 25%. Ion selection threshold was set to 1e4 counts. Isolation width of 3.0 Da was used.

Protein Identification and Label-free Quantitation MS raw files were submitted to MaxQuant software version 1.4.0.820 for protein identification using its Andromeda engine. Parameters were set as follow: protein N-acetylation and methionine oxidation as variable modifications. First search error window of 20 ppm and both precursor and MS/MS main search error set to 6 ppm. Trypsin without proline restriction enzyme option was used, with two allowed miscleavages. Minimal unique peptides were set to 1, and FDR allowed was 0.01 (1%) for peptide and protein identification. Label-free quantitation was set with a retention time alignment window of 3 min. The Uniprot human database was used (download from October 2014, with 85,915 entries). Generation of reversed sequences was selected to assign FDR rates. Known contaminants as provided by MaxQuant and identified in the samples were excluded from the analysis. The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium via the PRIDE21 partner repository with the dataset identifier PXD009302.

Western Blotting After indicated treatments, cells were lysed in lysis buffer supplemented with protease and phosphatase inhibitors or cells were eluted directly in sample buffer. The lysates were then separated by SDS-PAGE (4-20% gradient) and afterwards transferred to a PVDF membrane

9


Biochemistry 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

Page 10 of 30

(Bio-Rad) for western blotting. Blots were developed with SuperSignal West Dura Extended Chemiluminescent Substrate (Thermo Scientific) and detected using Gel Doc (Bio-Rad).

Co-immunoprecipitation and Streptavidin pulldown assays After indicated treatment, the cells were lysed in lysis buffer supplemented with protease inhibitors. Lysates were then centrifuged at 14 000 rpm for 10 minutes at 4°C. Supernatants were subjected to immunprecipitation reactions with indicated antibody immobilized to Dyneabeads Protein G or pulldown reactions with Streptavidin Sepharose. Protein complexes were then eluted in sample buffer, separated by SDS-PAGE and analyzed by western blotting.

Confocal microscopy For confocal microscopy, cells grown on coverslips were treated as indicated and fixed in 4% (w/v) formaldehyde (Sigma-Aldrich). The cells were permeabilized with 0.1 % (v/v) Triton X-100 in PBS, stained with indicated antibodies and mounted in mowiol. Cells stained with CD44 antibody were not permeabilized. Confocal images were acquired with a 63× objective on Zeiss LSM 780 and Zeiss LSM 710 confocal microscopes (Jena, Germany). Images were prepared with Zeiss LSM Image Browser.

10


Page 11 of 30 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

Biochemistry

RESULTS Construction of a functional FGF1 fusion protein for proximity labeling In order to produce a fusion protein for proximal biotin labeling, we fused a DNA fragment containing APEX2 to the 5’-end of the FGF1 cDNA in the expression vector pET3c (Figure 1A). To allow flexibility of the individual domains, we inserted a small (5 amino acids in length) linker between the two proteins. E. coli (BL21) transformed with this plasmid yielded abundant amounts of full-length fusion protein, which was purified by a heparin-Sepharose column (Figure 1B). Reassuringly, the fusion protein eluted from the heparin column at similar concentrations of NaCl as wild-type FGF1 (~1.1 M NaCl), indicating that the heparin binding affinity of FGF1 was retained in the fusion protein. Next, we tested whether APEX2 retained its peroxidase activity when fused with FGF1 and produced in E. coli. In a dot blot assay, APEX2-FGF1 was mixed with chemiluminescent substrate and chemiluminescence was detected with the use of a CCD camera. In the presence of the fusion protein, a strong signal was detected (Figure 1C), demonstrating that the peroxidase activity of bacterially produced APEX2 was highly functional in the fusion protein form. We also attempted to fuse the bacterial biotin ligase, BirA*22, to FGF1, but despite high protein purity, the activity of the ligase was low, making this approach ineffective (our unpublished results). We therefore proceeded with our APEX2-FGF1 fusion protein and next tested how it performed in interaction with cultured cells. First, we compared the ability of the fusion protein to activate FGFRs to that of wild-type FGF1. U2OS osteosarcoma cells stably expressing FGFR4 (U2OS-FGFR4) were incubated with FGF1 or APEX2-FGF1, and cell lysates were subjected to western blotting where activated FGFR4 was detected by anti-phospho-FGFR antibodies (Figure 1D). The fusion protein induced

11


Biochemistry 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

Page 12 of 30

activation of FGFR4 as efficiently as wild-type FGF1, indicating that it retains the ability to bind and activate FGFRs.

Figure 1. Characterization of APEX2-FGF1 (A) A schematic drawing of the APEX2-FGF1 construct. APEX2 (light grey) was genetically fused to the N-terminal end of FGF1 (dark grey). A sequence coding for a peptide linker (GlyGlyAlaGlyGly) was inserted between the two protein domains. (B) Recombinant APEX2-FGF1 protein expressed in E. coli was analyzed by SDSPAGE stained with Coomassie blue. (C) Peroxidase activity of APEX2-FGF1 was detected with luminol-based enhanced chemiluminescence HRP substrate and monitored with a CCD camera. (D) U2OS-FGFR4 cells were starved for 2 hours, then treated with 7 µM (100 ng/ml) FGF1 or 7 µM (300 ng/ml) APEX2-FGF1 in the presence of heparin (20 U/ml) for 30 minutes. Cells were

12


Page 13 of 30 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

Biochemistry

then lysed in sample buffer, separated by SDS-PAGE and analyzed by Western blotting with indicated antibodies. A p in front of the name of the antibody indicates that it recognizes the phosphorylated form of the protein.

The APEX2-FGF1 fusion protein efficiently biotinylates FGFR4. In order to test if the APEX2-FGF1 can be used to identify binding sites on cells, we incubated U2OS-FGFR4 cells with the fusion protein or with FGF1 as control. As an additional control, parental U2OS cells that are not stably expressing FGFR4 were treated with APEX2FGF1. The experiments were performed in the presence of heparin to promote binding to FGFRs. Cells were incubated with biotin-phenol and then briefly treated with H2O2 to induce biotin labeling14. Biotinylated proteins were then purified by streptavidin-Sepharose and analyzed by label-free, quantitative mass spectrometry (LC-MS/MS). As evident from the enrichment plots, FGFR4 was the most enriched candidate from this biotinylation screening (Figure 2A and Table S1), demonstrating the potential of the fusion protein to identify relevant binding sites on cell surfaces. Next, we validated the biotinylation of FGFR4 by western blotting. Cells were incubated with APEX2-FGF1 for different time points at 37°C. The cells were then lysed and biotinylated proteins in the lysate were purified by streptavidin-Sepharose. Western blots probed with antiFGFR4 antibodies confirmed that FGFR4 could be biotinylated and recovered by this method (Figure 2B). As expected, background labeling in the absence of biotin-phenol or H2O2 was low. Upon longer incubation periods with APEX2-FGF1, more FGFR4 was detected on the western blot, especially after incubation in the presence of the V-ATPase inhibitor Bafilomycin A1.

13


Biochemistry 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

Page 14 of 30

Although FGFR4 is mainly recycled upon FGF1 stimulation a fraction is also degraded23, likely accounting for the increased signal when degradation was blocked by Bafilomycin A1 treatment.

Figure 2. Efficient proximal biotin labeling by APEX2-FGF1. (A) Plots representing signal intensity versus enrichment. APEX2-FGF1 treated U2OS-FGFR4 cells (plot to the left) or FGF1 treated U2OS-FGFR4 cells (plot to the right) versus control (APEX2-FGF1 treated U2OS cells) are plotted against the corresponding signal intensity. (B) U2OS-FGFR4 cells were starved overnight in serum-free media. Then biotin-phenol (500 µM) was added to the cell culture

14


Page 15 of 30 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

Biochemistry

medium one hour before 100 ng/ml APEX2-FGF1 and 20 U/ml heparin were added for six hours (lane 4-5) or one hour (lane 1-3) at 37oC. As a control, one sample was not treated with biotinphenol (-Bio, lane 1). Moreover, one sample was treated with Bafilomycin A1 to prevent degradation in lysosomes (+ Baf, lane 5). At the end of APEX2-FGF1 incubation, H2O2 (1 mM) was added for 1 minute to initiate biotinylation. As a control, one sample was not treated with H2O2 (lane 2). The reaction was quenched and the cells were lysed. The lysates were subjected to Streptavidin Sepharose pulldown followed by SDS-PAGE and western blotting with indicated antibody.

Identification of novel proteoglycan binding sites for FGF1 Encouraged by the result that APEX2-FGF1 fusion protein efficiently biotinylated the FGF1 receptor, FGFR4, we proceeded to test whether our approach could identify novel interactions between FGF1 and cell surface proteoglycans. We therefore incubated RPE1 cells with APEX2-FGF1 in the presence or absence of heparin at 4°C (Figure 3A). Heparin binds strongly to FGF1 and competes for binding to cell surface proteoglycans (Figure 3B). By comparing the two conditions we wished to identify proteoglycans that specifically bound to the positively charged heparin-binding sites of FGF1. In cells treated at 4°C with APEX2-FGF1 in the absence of heparin, we observed a smear of bands indicating efficient biotin labeling (Figure 3C, lane 3). This labeling was efficiently reduced in the presence of heparin (Figure 3C, lane 2), indicating that the biotinylation was dependent on binding of APEX2-FGF1 to proteoglycans on the cell-surface. Next, we used quantitative mass spectrometry (LC-MS/MS) to identify the proteins that were outcompeted with heparin. RPE1 cells were incubated with biotin-phenol before APEX2-

15


Biochemistry 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

Page 16 of 30

FGF1 or FGF1 were added for 2 hours at 4°C to allow binding to the cell-surface. The cells were then washed to remove excess APEX2-FGF1/FGF1 before initiation of the biotinylation process by addition of H2O2. Biotinylated proteins were purified by streptavidin-Sepharose and analyzed by quantitative MS. Several proteins were highly enriched in our samples compared to control (Figure 3D). These proteins were also outcompeted by heparin, indicating that the interactions were dependent on the heparin-binding part of FGF1. The two most enriched candidates (enriched more than 4000 times) were CSPG4 and CD44 (Table 1 and Table S2). Importantly, these proteins are plausible candidate FGF1 interactors as both contain negatively charged glycosaminoglycan chains. CD44 is facultative and can possess GAG chains depending on its splice form15. The negative charge of these glycan chains is likely to mediate binding to the positively charged heparin binding sites of FGF1. The most enriched candidates (Table 1) were all cell surface membrane proteins. Apart from the two top candidates (CD44 and CSPG4), the other candidates are not expected to contain GAGs even though their labeling was competed out with heparin. Possibly, FGF1-APEX2 binds to proteoglycans and these other hits represent indirect tagging of proteins in the molecular neighborhood of the of the APEX domain. In line with this reasoning, 5 of our top candidates have previously been shown to interact with CD44 (Figure 3E). Altogether we found 125 proteins that were significantly (p

Proximity labeling by a recombinant APEX2-FGF1 fusion protein

Recommend Documents