Identification of Siglec Ligands Using a Proximity ... - ACS Publications

Subscriber access provided by University of Sussex Library

Article

Identification of Siglec ligands using a proximity labeling method Lanyi Chang, Yi-Ju Chen, Chan-Yo Fan, Chin-Ju Tang, Yi-Hsiu Chen, PenkYeir Low, Albert Ventura, Chun-Cheng Lin, Yu-Ju Chen, and Takashi Angata J. Proteome Res., Just Accepted Manuscript • DOI: 10.1021/acs.jproteome.7b00625 • Publication Date (Web): 13 Sep 2017 Downloaded from http://pubs.acs.org on September 16, 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 46

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

Identification of Siglec ligands using a proximity labeling method Lanyi Chang†,¶, Yi-Ju Chen‡,¶, Chan-Yo Fan§, Chin-Ju Tang†, Yi-Hsiu Chen†, Penk-Yeir Low†, Albert Ventura†, Chun-Cheng Lin§, Yu-Ju Chen‡, and Takashi Angata†,⊥,*

†

Institute of Biological Chemistry and ‡ Institute of Chemistry, Academia Sinica, Taipei 115,

Taiwan; § Department of Chemistry, National Tsing Hua University, Hsinchu 300, Taiwan; ⊥ Institute of Biochemical Sciences, National Taiwan University, Taipei 106, Taiwan.

KEYWORDS Siglec, sialic acid, lectin, ligand, tyramide, peroxidase, proximity labeling, proteomics

ACS Paragon Plus Environment

1


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 46

ABSTRACT

Siglecs are a family of receptor-type glycan recognition proteins (lectins) involved in selfnonself discrimination by the immune system. Identification of Siglec ligands is necessary to understand how Siglec-ligand interaction translates into biological outcomes. However, this is challenging because the interaction is weak. To facilitate identification of Siglec ligands, we adopted a proximity labeling method based on the tyramide radicalization principle. Cells that express Siglec ligands were labeled with Siglec-peroxidase complexes and incubated with biotin tyramide and hydrogen peroxide, to generate short-lived tyramide radicals that covalently label the proteins near the Siglec-peroxidase complex. A proof-of-principle experiment using CD22 (Siglec-2) probe identified its known ligands on B cells, including CD22 itself, CD45, and IgM among others, demonstrating the validity of this method. The specificity of labeling was confirmed by sialidase treatment of target cells and using glycan recognition-deficient mutant CD22 probes. Moreover, possible interactions between biotin-labeled proteins were revealed by literature-based protein-protein interaction network analysis, implying the presence of a molecular cluster comprising CD22 ligands. Further application of this method identified CD44 as a hitherto unknown Siglec-15 ligand on RAW264.7-derived osteoclasts. These results demonstrated the utility of proximity labeling for the identification of Siglec ligands, which may extend to other lectins.


2

Page 3 of 46

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


INTRODUCTION

Specific recognition of glycans by endogenous glycan-recognition proteins (lectins) is essential for proper regulation of cell-to-cell communication, such as leukocyte recruitment to the site of inflammation mediated by selectins and their ligands.1 Siglecs are a family of vertebrate lectins involved in immune cell recognition and cell-to-cell interaction.2-4 Siglec-ligand interactions also promote tolerance toward self-antigens.5-6 Recent studies have shown that some Siglecs recognize tumor-associated glycans, which generally favor survival or spread of tumor cells.7-10 Identification of Siglec ligands would allow us to understand the molecular mechanisms of Siglec-ligand interaction and its biological consequences in greater detail, which may eventually lead to a strategy to target such interactions for therapeutic purposes. However, identification of Siglec ligands remains technically challenging because of the inherently weak interactions between Siglecs and their ligands. Several studies have identified Siglec ligands by conventional affinity purification,11-14 by chemical functionalization of ligands by metabolic engineering,15-16 or by a candidate-based approach.17 However, each of these approaches has shortcomings. For example, conventional affinity purification requires a relatively large amount of recombinant Siglec for the preparation of affinity media and the starting material (cells/tissues) from which the ligand shall be purified. Also, identification of membrane-integral glycoproteins such as Siglec ligands requires detergent-mediated cell/tissue lysis, which destroys higher-order protein assembly that may be essential for Siglec-ligand interaction. Chemical functionalization of sialic acids (for example by the addition of a photo-reactive group) by metabolic labeling overcomes this limitation,15-16, 18 but requires detailed knowledge of the binding specificity of the Siglec of interest, so that the


3


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 46

functional group does not interfere with Siglec binding activity. In addition, the metabolic precursor of such chemically functionalized sialic acid must compete against endogenous precursors of sialic acid to be displayed on the cell surface, potentially limiting the efficiency of cell-surface expression.19 Hypothesis-based ligand identification, by definition, requires a short list of ligand candidates, which is obviously not available a priori. A bridging method to facilitate the identification of Siglec ligand candidates in a hypothesis-free manner will be of great help to identify biologically meaningful ligands by a candidate-based approach. Proximity labeling methods,20-23 utilizing a bait protein fused (or in complex) with an enzyme that generates short-lived radical species to label proteins in its vicinity (typically with biotin), have gained popularity as ways to identify constituents of protein complexes. In particular,

tyramide

radicalization,24

which

has

been

utilized

for

decades

in

immunohistochemistry to amplify antibody binding or nucleotide hybridization signals,25-27 is now gaining popularity as a chemical principle of proximity labeling for its relatively high resolution (i.e., small diffusion radius of radicals generated), ease of use, and versatility. In this study, we adopted a proximity labeling method based on the tyramide radicalization principle to identify Siglec ligand candidates in a hypothesis-free manner, to facilitate subsequent hypothesis-driven discovery of biologically relevant Siglec ligands. As a proof of principle, we targeted CD22 (Siglec-2) ligands on a B lymphoma cell line. The validity of our method was verified by the significant overlap between the CD22/Siglec-2 ligand candidates identified by our method and those reported by a previous study that employed a chemical biology approach to systematically identify such candidates.16 Our method was further applied to identify biologically relevant ligand candidates for Siglec-15, a Siglec involved in the


4

Page 5 of 46

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


fusion of osteoclast precursors by interacting with a hitherto unidentified ligand, demonstrating the utility of the method to discover Siglec ligands in an exploratory setting.

EXPERIMENTAL PROCEDURES

Synthesis of photo-cleavable biotin tyramide The synthesis of photo-cleavable biotin-tyramide (compound 7) is described in detail in the Supplementary Experimental Procedures.

Cell culture BJAB (a human B lymphoid cell line) was kindly provided by Dr. Kuo-I Lin (Genomics Research Center, Academia Sinica), and maintained in RPMI1640 supplemented with 20% fetal bovine serum (FBS) and penicillin-streptomycin. RAW264.7 cells were obtained from the American Type Culture Collection (Manassas, VA, USA) and maintained in DMEM supplemented with 10% FBS and penicillin-streptomycin. Osteoclast differentiation of RAW264.7 cells was induced by culturing the cells in the presence of mouse receptor activator of nuclear factor kappa B ligand (RANKL; 40 ng/ml) for 3 to 5 days.28

Preparation of Siglec-Fc fusion proteins Recombinant fusion protein, consisting of the three N-terminal immunoglobulin-like domains of CD22/Siglec-2 (amino acid #1–330) followed by a FLAG tag and Fc fragment of human IgG1, was prepared as described previously,29 with modifications. Briefly, the protein expression


5


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 46

construct (Siglec-2-AvT-EK-Fc/pcDNA3.1) was transfected into Expi293F cells (ThermoFisher Scientific, Waltham, MA, USA) using ExpiFectamine 293 Transfection Kit (ThermoFisher Scientific) in accordance with the manufacturer's instructions. The cells were cultured for 7 days, with a medium change 4 days after transfection. Culture supernatant was cleared of cell debris by centrifugation at 2,000 g for 10 min. Cleared culture supernatant was mixed with rProtein ASepharose Fast Flow (GE Healthcare, Piscataway, NJ, USA) and incubated overnight at 4˚C. The media was packed into a disposable chromatography column (Poly-Prep Chromatography Columns, Bio-Rad, Hercules, CA, USA), washed with Dulbecco's phosphate buffered saline (DPBS), treated in column with Arthrobacter ureafaciens sialidase (Nacalai Tesque, Kyoto, Japan) to remove sialic acids, and washed again with D-PBS. Recombinant Siglec-2-Fc protein was then eluted with 0.1 M sodium citrate buffer, pH 3.0, and neutralized with 1/10 volume of 1 M sodium carbonate. The neutralized eluate was concentrated with an UltraFree ultrafiltration device (Millipore, Billerica, MA, USA) and buffer-exchanged to D-PBS. The plasmid construct for the expression of a mutant Siglec-2[R120A]-Fc fusion protein, which is deficient in sialic acid binding (with the Arg120 residue required for sialic acid binding replaced by Ala), was prepared by site-directed mutagenesis of the Siglec-2-Fc fusion protein expression construct mentioned above using the Q5 Site-Directed Mutagenesis Kit (New England Biolabs, Ipswich, MA, USA). The mutated protein was produced as described above. Mouse Siglec-15-Fc and mutant mouse Siglec-15[R143A]-Fc deficient in sialic acid binding were prepared in a similar manner, as described previously.30

Biotin labeling of Siglec ligand candidates by biotin tyramide radicalization


6

Page 7 of 46

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


Recombinant Siglec-Fc fusion protein (10 µg) was incubated with M2-peroxidase antibody (5 µg; an anti-FLAG monoclonal antibody directly conjugated to horseradish peroxidase (HRP), Sigma-Aldrich, St Louis, MO, USA) on ice for 30 to 60 minutes to facilitate complex formation. Cells (20 × 106 cells/reaction) were incubated on ice for 1 h with the Siglec-HRP complex prepared above. The cells were washed twice with 10 ml of 140 mM NaCl in 20 mM Tris-HCl buffer, pH 8.0 (TBS), then incubated with 10 µM of compound 7 in 10 mM H2O2 in TBS at room temperature for 10 min. The cells were washed three times with 10 ml TBS, and lysed in 200 µl of lysis buffer (50 mM Tris-HCl buffer, pH 8.0, 150 mM NaCl, 1% Nonidet P-40, 0.5% sodium deoxicholate, 0.1% sodium dodecyl sulfate (SDS), supplemented with protease inhibitor cocktail (cOmplete ULTRA Tablets Mini, EDTA-free; Roche, Basel, Switzerland)) on ice for 30 min. The lysate was cleared by centrifugation (15,000 g, 30 min), and immediately used for purification of biotinylated proteins (as described below) or stored at -30˚C until use. For comparison, cell-surface proteins on BJAB cells were labeled with sulfo-NHS-biotin (2.3 mM in D-PBS, on ice for 1 h), which covalently introduces biotin into primary amine groups, in accordance with the protocol provided by the manufacturer (ThermoFisher Scientific).

Purification of biotinylated proteins To purify biotinylated proteins, we used either streptavidin-functionalized paramagnetic beads or agarose resin. Purification with paramagnetic beads: One mg of streptavidin-functionalized paramagnetic beads (Dynabeads Streptavidin-MyOne C1; ThermoFisher Scientific) was mixed with 100 µl of cleared cell lysate, and incubated at room temperature for 1 h with constant shaking (600 rpm, on a horizontal shaker). The beads were collected by magnet (DynaMag-2, ThermoFisher


7


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 46

Scientific), and washed extensively with D-PBS containing 0.1% SDS. Captured proteins were eluted from the beads by boiling at 100˚C for 5 min in 50 µl of 1× SDS-polyacrylamide gel electrophoresis (PAGE) sample buffer (Bio-Rad) containing 100 mM dithiothreitol (DTT). Alternatively, the proteins were eluted by photo-cleavage of the linker in biotin-tyramide by exposure to ultraviolet light (2 × 1 min exposure to 365 nm light; model LC-L1 V3 equipped with collimator-type LED head unit, output approximately 1,000 mW/cm2; Hamamatsu Photonics, Hamamatsu, Japan) in 50 mM ammonium bicarbonate containing 0.1% RapiGest SF (Waters, Milford, MA, USA), and lyophilized to remove salt. The eluted proteins were subjected to protein identification by mass spectrometry or SDS-PAGE/western blot analysis, as described below. Purification with agarose beads: Two hundred and fifty µl of Streptavidin Sepharose High Performance (GE Healthcare Life Sciences) was mixed with 250 µl of cleared lysate, and incubated at room temperature with gentle agitation (40 rpm, on a rocking shaker). The mixture was centrifuged (1,000 g, 30 sec) to remove the supernatant, and the beads were washed three times with 1 ml D-PBS containing 0.1% SDS. The beads were aliquoted into two tubes (125 µl each). One tube received 25 µl of 6× SDS-PAGE sample buffer containing 600 mM DTT, was boiled at 100˚C for 5 min, and then centrifuged to obtain the supernatant (approximately 25 µl). The other tube received 100 µl of 50 mM ammonium bicarbonate containing 0.1% RapiGest SF, was irradiated with a 365 nm UV lamp (trans-illuminator placed upside-down on the tube with the open cap; 365 nm, 8 W capacity, model LM-26, UVP) for 20 min to cleave the linker in biotin-tyramide, and was centrifuged to obtain the supernatant (approximately 100 µl). The supernatant was lyophilized and re-suspended in 20 µl of 1× SDS-PAGE sample buffer


8

Page 9 of 46

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


containing 100 mM DTT, and boiled at 100˚C for 5 min. The eluted proteins were subjected to protein identification by mass spectrometry, as described below.

In-gel digestion of purified proteins and mass spectrometry (MS) analysis The purified biotinylated proteins were separated by short SDS-PAGE (approximately 0.5 cm into separation gel). The gel area containing proteins was excised, cut into smaller pieces, destained with 50% acetonitrile (ACN) plus 25 mM triethylammonium bicarbonate (TEABC), and subjected to in-gel reduction and digestion. The proteins in the gel pieces were reduced with tris(2-carboxyethyl)phosphine (TCEP) at a final concentration of 5 mM at 37˚C for 30 min in the dark, and alkylated with iodoacetamide (IAM) at a final concentration of 20 mM at 37˚C for 60 min in the dark. After extraction with 100% ACN and removing all liquid, the gel pieces were re-saturated with 25 mM TEABC, to which 2 µg trypsin was added, and were incubated at 37˚C for 16 h. Digested peptides were extracted with 50% ACN/5% formic acid (FA), twice, and 100% ACN and dried completely under vacuum. The peptides were desalted by C18 Zip-tip (Millipore) and subjected to Orbitrap Fusion Tribrid Mass Spectrometer (ThermoFisher Scientific, San Jose, CA) equipped with a PicoView nanospray interface (New Objective, Woburn, MA, USA). Peptides were loaded onto an analytical C18 column (Acclaim PepMap RSLC, 75 µm i.d. × 25 cm length; ThermoFisher Scientific) packed with 2 µm particles with a pore size of 100 Å, and were separated using a segmented gradient for 120 min with the following mobile phases: water with 0.1% FA (buffer A) and 2% to 85% ACN with 0.1% FA (buffer B) at 500 nL/min flow rate. Survey scans of peptide precursors from 400 to 1650 m/z with charge states 2–5 were performed at 60K resolution and the AGC target was set to 3 × 105 by Orbitrap. Tandem MS was performed by isolation window at 1.6 Da with the quadrupole,


9


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 46

AGC target set to 1 × 105, and used for higher-energy collisional dissociation (HCD) fragmentation detected in Orbitrap at a resolution setting of 15K with a normalized collision energy (NCE) of 27%. Top 15 of given precursors were selectively fragmented and scanned out in mass spectra.

Protein identification and quantitation The raw data were processed using Proteome Discoverer 2.1 (PD2.1; ThermoFisher Scientific), and peptide identification was performed by Mascot search engine (version 2.3.2) and SEQUEST search engine against the Swiss-Prot database (v2015_12, total 20,193 sequences from human) with a percolator (strict false discovery rate (FDR) of 0.01 and a relaxed FDR of 0.05). The protease was specified as trypsin with 2 maximum missing cleavage sites. Mass tolerance for precursor ion mass was 10 ppm with the fragment ion tolerance as 0.1 Da. Methionine oxidation, cysteine alkylation by iodoacetamide (carbamidomethyl), and deamidation at asparagine or glutamine, were selected as variable modifications. Peptides were considered identified if their individual ion score was higher than the identity score (p < 0.05). To evaluate the false discovery rate (< 1%) in protein identification, a decoy database search against a randomized decoy database created by PD2.1 using identical search parameters and validation criteria was also performed. Peptide-spectrum matches (PSMs) with at least high confidence and a strict maximum parsimony principle (target FDR

Identification of Siglec Ligands Using a Proximity ... - ACS Publications

Recommend Documents