- Email: [email protected]
Identification of human plasma C1 inhibitor as a target protein for staphylococcal superantigen-like protein 5 (SSL5)
Biochemical and Biophysical Research Communications xxx (xxxx) xxx
Contents lists available at ScienceDirect
Biochemical and Biophysical Research Communications journal homepage: www.elsevier.com/locate/ybbrc
Identification of human plasma C1 inhibitor as a target protein for staphylococcal superantigen-like protein 5 (SSL5) Teruaki Oku*, Chisato Kurisaka, Yusuke Ando, Tsutomu Tsuji** Department of Microbiology, Hoshi University School of Pharmacy and Pharmaceutical Sciences, Tokyo, Japan
a r t i c l e i n f o
a b s t r a c t
Article history: Received 27 November 2018 Accepted 5 December 2018 Available online xxx
The family of staphylococcal superantigen-like proteins (SSLs) have a structure similar to bacterial superantigens but exhibit no superantigenic activity. These exoproteins have recently been shown to disturb the host immune defense system. One family member, SSL5, was reported to bind to human leukocyte P-selectin glycoprotein ligand-1 (PSGL-1) and matrix metalloproteinase-9 (MMP-9) and to interfere with leukocyte trafficking. In the present study, we explored human plasma proteins bound by glutathione S-transferase (GST)-tagged recombinant SSL5 (GST-SSL5) and identified plasma protease C1 inhibitor (C1Inh) as a major SSL5-binding protein based on the results of peptide mass fingerprinting analysis with MALDI-TOFMS. GST-SSL5 was found to attenuate the inhibitory activity of recombinant histidine-tagged C1Inh (C1Inh-His) toward complement C1s. We also observed that the treatment of C1Inh-His with neuraminidase markedly decreased its binding to GST-SSL5. Moreover, C1Inh-His produced by Lec2 mutant cells (deficient in sialic acid biosynthesis) showed much lower binding affinity for SSL5 than that produced by the wild-type CHO-K1 cells, as assessed by pull-down assay. These results suggest that SSL5 binds to C1Inh in a sialic acid-dependent fashion and modulates the host immune defense through perturbation of the complement system in association with S. aureus infection. © 2018 Elsevier Inc. All rights reserved.
Keywords: Staphylococcus aureus SSL5 C1 inhibitor Complement Immune perturbation Sialic acid
1. Introduction Staphylococcus aureus is Gram-positive bacterium, and its infection is considered to be the most frequent cause of a variety of diseases ranging from mild skin infection to serious infection such as sepsis [1]. S. aureus secretes various proteins that bind to critical molecules involved in the host innate and adaptive immune system as well as cytotoxins [2,3]. Staphylococcal superantigen-like proteins (SSLs) are a family of exoproteins that have a structure similar to bacterial superantigens such as toxic shock syndrome toxine-1 (TSST-1) and streptococcal pyrogenic exotoxin C (SPEC) but have no superantigenic activity because they lack a T cell receptor (TCR)-binding domain [4]. The family is composed of 14 members
Abbreviations: SSL, staphylococcal superantigen-like protein; GST, glutathione S-transferase; C1Inh, plasma protease C1 inhibitor; MMP-9, matrix metalloproteinase-9; PBS, phosphate-buffered saline; SDS-PAGE, sodium dodecyl sulfatepolyacrylamide gel electrophoresis; MALDI-TOFMS, matrix-assisted laser desorption/ionization-time of flight mass spectrometry; HRP, horseradish peroxidase; DTNB 5, 5’-dithiobis(2-nitrobenzoic acid); PNGase F, peptide-N-glycosidase F. * Corresponding author. ** Corresponding author. E-mail addresses: [email protected] (T. Oku), [email protected] (T. Tsuji).
(SSL1-SSL14), and these proteins possess a common structure, the N-terminal b-barrel globular domain (OB-fold) and C-terminal bgrasp domain [4,5]. Previous studies have indicated that SSL5 binds specifically to P-selectin glycoprotein ligand-1 (PSGL-1) on leukocyte cell membranes [6] and to platelet membrane glycoproteins (GPIb and GPVI) [7]. We also reported that SSL5 bound to matrix metalloproteinase-9 (MMP-9) from human neutrophils and inhibited its enzymatic activity [8]. These results strongly suggest that SSL5 has significant influences on leukocyte trafficking and platelet activation. Because PSGL-1, GPIb, GPVI and MMP-9 are glycoproteins with rather high carbohydrate contents, the carbohydrate chains of these glycoproteins are thought to be important for their interactions with SSL5. Indeed, we found that the interaction between SSL5 and MMP-9 involved sialylated O-glycans of MMP-9 [9]. Coagulase produced by S. aureus is a well-known exoenzyme that binds to plasma prothrombin to form a complex with catalytic activity for the conversion of fibrinogen to fibrin. In addition, it has been reported that S. aureus secretes several exoproteins reacting with plasma proteins, including blood coagulation factors and complement components [10e12]. Some SSL proteins were also shown to bind to plasma proteins; e.g., SSL7 bound to human IgA
https://doi.org/10.1016/j.bbrc.2018.12.026 0006-291X/© 2018 Elsevier Inc. All rights reserved.
Please cite this article as: T. Oku et al., Identification of human plasma C1 inhibitor as a target protein for staphylococcal superantigen-like protein 5 (SSL5), Biochemical and Biophysical Research Communications, https://doi.org/10.1016/j.bbrc.2018.12.026
2
T. Oku et al. / Biochemical and Biophysical Research Communications xxx (xxxx) xxx
and complement C5 [13e15], and SSL10 bound to human IgG, prothrombin and factor Xa [16e18]. Because SSLs suppressed phagocytosis by leukocytes, blood coagulation and complement activation through these interactions, SSLs were considered to play roles in the immune evasion of S. aureus. In this study, therefore, we attempted to explore possible target plasma proteins for SSL5 by using affinity purification, and found that recombinant SSL5 specifically bound to a complement regulator. 2. Materials and methods 2.1. Reagents Isopropyl-b-D-thiogalactopyranoside (IPTG), glutathione, and imidazole were obtained from FUJIFILM Wako Pure Chemical Corp. (Osaka, Japan). Brij 35, RPMI1640 medium, trypsin, 5, 5’-dithiobis(2-nitrobenzoic acid) (DTNB), Triton X-100, and TriReagent were purchased from Sigma-Aldrich (St. Louis, MO, USA). One step Coomassie Brilliant Blue (CBB) was purchased from Bio Craft (Tokyo, Japan). PrimeSTAR MAX DNA polymerase was obtained from Takara Bio Inc. (Shiga, Japan). Oligonucleotides were supplied by FASMAC (Kanagawa, Japan). cOmplete™ His-Tag Purification Resin and 1,4-dithiothreitol (DTT) were purchased from Roche Diagnostics (Indianapolis, IN, USA). Glutathione-agarose was obtained from Thermo Fisher Scientific (Waltham, MA, USA). Human pooled plasma and fetal calf serum (FCS) were supplied by Kohjin Bio (Saitama, Japan) and Biosera (Boussens, France), respectively. A monoclonal antibody against glutathione S-transferase (GST) was prepared in our laboratory [19]. Anti-6 His mouse monoclonal antibody and horseradish peroxidase (HRP)-conjugated goat antimouse IgG antibody were purchased from Thermo Fisher Scientific and Kirkegaard & Perry Laboratories (Gaithersburg, MD, USA), respectively. Biotin-conjugated Maackia amurensis mitogen (MAM) was purchased from Cosmo Bio Co., Ltd. (Tokyo, Japan). HRPconjugated streptavidin was obtained from BioLegend (San Diego, CA, USA). Peptide-N-glycosidase F (PNGase F) and neuraminidase (Salmonella typhimurium LT2) were purchased from New England Biolabs (Ipswich, MA, USA).
0.25% Brij 35 at 4 C for 1 h. After the beads were washed with the same buffer, proteins bound to the beads were recovered with SDS sample buffer (50 mM Tris-HCl, 1% SDS, 5% glycerol, 0.01% bromophenol blue, pH 6.8) and analyzed by SDS-PAGE and silver staining. 2.4. In-gel digestion and peptide mass fingerprinting (PMF) The protein band was excised from the SDS-PAGE gel stained with silver nitrate, destained with 15 mM potassium ferricyanide and 50 mM sodium thiosulfate, reduced with 10 mM DTT in 25 mM ammonium bicarbonate, and alkylated with 55 mM iodoacetamide in 25 mM ammonium bicarbonate in a water/acetonitrile (50/50) mixture. The reduced/alkylated protein was digested by treatment of the gel with trypsin (10 mg/mL) in 45 mM ammonium bicarbonate. The digested peptides were extracted with 5% trifluoroacetic acid in 50% acetonitrile. Samples were concentrated under reduced pressure and purified with C18 ZipTips (Millipore, Billerica, MA, USA) following the manufacturer's instructions. Peptide mass fingerprinting analysis was performed using MALDI-TOFMS (AXIMA-QIT; Shimadzu, Kyoto, Japan), and monoisotopic masses were processed for identification through analysis with m/z software and a search of the SwissProt database using MASCOT software (Matrix Science, Boston, MA, USA). 2.5. Preparation of recombinant C1 inhibitor (C1Inh-His)
The SSL5 gene was inserted into pGEX-5X-1 vector (GE Healthcare, Piscataway, NJ, USA) as described previously [9]. The recombinant GST-tagged SSL5 (GST-SSL5) was expressed in Escherichia coli (SoluBL21; Genlantis, San Diego, CA, USA), and purified as described previously [20,21]. Briefly, the transformed E. coli grown to exponential phase was cultured further in TB medium (12 g/L tryptone, 24 g/L yeast extract, 0.4% (v/v) glycerol, 17 mM KH2PO4 and 72 mM K2HPO4) containing 0.2 mM IPTG at 25 C for 16 h. The cells harvested by centrifugation were resuspended in sonication buffer (50 mM Tris/HCl, pH 8.0, containing 150 mM NaCl and 1 mM EDTA), and lysozyme (final 1.0 mg/mL) was added to the suspension. The cells were disrupted by sonication and centrifuged. The supernatant was mixed with glutathione agarose (Thermo Fisher Scientific) and incubated at 4 C for 1 h with gentle agitation. After washing with sonication buffer, GST fusion protein was eluted with 10 mM of glutathione (reduced form). Purified protein was applied to a PD-10 column (GE Healthcare) to replace the buffer with phosphate-buffered saline (PBS).
A DNA construct for the expression of a fusion protein of human plasma protease C1 inhibitor with His-tag (C1Inh-His) with a flexible spacer (Gly4-Ser) [22] at the C-terminus of C1Inh (C1IN) was amplified by PCR using cDNA derived from the human mononuclear leukocytes as a template. PCR was conducted with the following set of primers: 50 -AGC GGC CGC CAC CAT GGC CTC CAG GCT G-3’ (sense) and 50 -CTC GAG TCA ATG GTG ATG GTG ATG ATG GGA TCC ACC ACC TCC GGC CCT GGG GTC ATA TAC TCG-3’ (antisense). The product was cloned into the NotI/XhoI restriction site of pcDNA6-V5HisA (Thermo Fisher Scientific) (pcDNA6-C1Inh-His). The full-length cDNA sequence of the C1Inh-His was determined by the dideoxy method. The plasmid (pcDNA6-C1Inh-His) was introduced into HEK293FT (Thermo Fisher Scientific), CHO-K1 or Lec2 cells (CRL1736; American Type Culture Collection) with polyethylenimine MAX (PEI) (Mw 40,000) as described previously [23]. The culture supernatant (500 mL) was mixed with cOmplete His-tag purification resin (Roche Diagnostics) and incubated 1 h at 4 C with gentle agitation. The resin was washed with 50 mM Tris-HCl, 500 mM NaCl, and 15 mM imidazole (pH 7.4), and the C1Inh-His bound to the resin was eluted with 50 mM Tris-HCl, 500 mM NaCl, and 200 mM imidazole (pH 7.4). Purified C1Inh-His was applied to a PD10 column to replace the buffer with PBS. Inhibitory activity of C1Inh-His against C1 esterase was measured by a functional chromogenic assay (SIEMENS, Eschborn, Germany) according to the manufacturer's instructions. Briefly, purified C1Inh-His (2 mL, 1 mg/mL) or human plasma was treated with C1 esterase (0.1 mL) at 37 C for 5 min. The mixture was incubated with a synthetic substrate for C1 esterase (MeOC-Lys(εCbo)Gly-Arg-pNA, 5 mM) (0.01 mL) at 37 C for 5 min. The reaction was stopped with 20% acetic acid (0.05 mL), and absorbance at 415 nm was measured with a microplate reader (MTP-450; Corona, Tokyo, Japan).
2.3. Pull-down assay
2.6. Measurement of C1s activity
Purified GST-SSL5 or GST (5 mg) was bound to glutathione agarose (30 mL, 50% slurry) at 4 C for 1 h. The agarose beads were incubated with human plasma (100 mL) diluted with PBS containing
Complement C1s activity was measured using a chromogenic substrate, N-carbobenzyloxy-Lys-thiobenzyl ester (Bachem, Bubendorf, Switzerland), and DTNB. The assay was performed
2.2. Preparation of recombinant GST-SSL5
Please cite this article as: T. Oku et al., Identification of human plasma C1 inhibitor as a target protein for staphylococcal superantigen-like protein 5 (SSL5), Biochemical and Biophysical Research Communications, https://doi.org/10.1016/j.bbrc.2018.12.026
T. Oku et al. / Biochemical and Biophysical Research Communications xxx (xxxx) xxx
according to the manufacturer's instructions using recombinant complement C1s (Merck, Bedford, MA, USA) as a positive control. Experiments were performed in triplicate, and statistical data analysis was conducted using the Student's t-test. 2.7. Treatments with PNGase F and neuraminidase N-glycans of purified C1Inh-His (1 mg) were removed by incubation with PNGase F (500 units) in 50 mM sodium phosphate buffer (pH 6.0) at 37 C for 3 h. Sialic acid residues were hydrolyzed with neuraminidase (50 units) in 50 mM sodium acetate buffer (pH 5.5) containing 5 mM CaCl2 at 37 C for 3 h. 3. Results 3.1. Recombinant GST-SSL5 binds to C1 inhibitor from human plasma The purified recombinant GST-SSL5 and GST were analyzed by SDS-PAGE, and each protein gave an apparently single band at ~48 kDa (GST-SSL5) or ~25 kDa (GST) after staining with CBB (Fig. 1A). We attempted to isolate SSL5-binding proteins from human plasma by pull-down assay. When the proteins bound by GSTSSL5 were analyzed by SDS-PAGE and silver staining, a major protein band of ~100 kDa was detected (Fig. 1B). This protein was analyzed by peptide mass fingerprinting with MALDI-TOFMS and identified as plasma protease C1 inhibitor (C1Inh; also called C1 inhibitor, C1 esterase inhibitor and complement C1 inactivator) by the MASCOT search engine. The mass peaks of amino acid sequences corresponding to human C1 inhibitor (UniProt: P05155) detected by MS were as follows: m/z 853.45 for 488FPVFMGR494, m/z 910.48 for 242TLYSSSPR249, m/z 935.46 for 310MEPFHFK316, m/z 1067.53 for 153LYHAFSAMK161, m/z 1116.59 for 277LLDSLPSDTR286, 391 m/z 1185.70 for FQPTLLTLPR400, m/z 1218.59 for 202 211 DFTCVHQALK , m/z 1264.68 for 191TNLESILSYPK201, m/z 1317.80 for 287LVLLNAIYLSAK298, m/z 1431.73 for 330YPVAHFIDQTLK341, m/z 1482.75 for 403VTTSQDMLSIMEK415, m/z 1559.87 for 329KYPVAHFIDQTLK341, m/z 1593.79 for 367LEDMEQALSPSVFK380, m/z 1826.98
217 for GVTSVSQIFHSPDLAIR233 HRLEDMEQALSPSVFK380.
3
and
m/z
1886.97
for
365
3.2. SSL5 binds to recombinant C1 inhibitor To further examine the interaction between SSL5 and C1 inhibitor, we prepared a recombinant histidine-tagged C1 inhibitor (C1Inh-His) (Fig. 2A). C1Inh-His expressed in HEK293FT cells was purified from the culture supernatants with His-tag purification resin. SDS-PAGE analysis of the purified protein gave an apparently single band at ~100 kDa after staining with CBB (Fig. 2B). The inhibitory activity of C1Inh-His against C1 esterase was measured using a synthetic substrate. The addition of purified C1Inh-His (18 mg/mL) decreased the esterase activity to approximately 45%, a level comparable to that in human plasma (1.8%) (Fig. 2C). Next, we performed a pull-down assay using immobilized GST-SSL5 and soluble C1Inh-His followed by SDS-PAGE and CBB staining. As shown in Fig. 2D, the fraction bound to GST-SSL5-agarose gave the protein band corresponding to C1Inh-His (~100 kDa), whereas no band was detected from the fraction bound to GST-agarose. Conversely, the pull-down assay using immobilized C1Inh-His and soluble GST-SSL5 resulted in the binding of GST-SSL5 but not GST to immobilized C1Inh-His (Fig. 2E). 3.3. SSL5 attenuates C1 inhibitor activity C1 inhibitor shows inhibitory activity against complement components C1s and C1r and thus regulates complement activation. We also confirmed that C1Inh-His inhibited C1s activity in a dose-dependent manner as measured with a colorimetric peptide substrate, N-carbobenzyloxy-Lys-thiobenzyl ester (Z-K-SBzl) (Fig. 3A). We next examined the effect of GST-SSL5 on the inhibitory activity of C1Inh-His toward C1s. As shown in Fig. 3B, C1s activity inhibited in the presence of C1Inh-His was partially recovered by the preincubation with GST-SSL5 but not with control GST; i.e., the 66% inhibition observed in the presence of C1Inh-His (25 mg/mL) was reduced to 28% inhibition after the addition of GST-SSL5 (400 mg/mL). 3.4. Sialic acid-containing glycans of C1 inhibitor are involved in the interaction with SSL5
Fig. 1. Isolation of SSL5-binding proteins from human plasma. (A) Purified recombinant GST-SSL5 and the control GST were separated by SDS-PAGE (5e20% gradient gel) and stained with CBB. (B) Human plasma was incubated with GST-SSL5agarose or control GST-agarose, and the bound proteins were separated by SDS-PAGE (7.5%) followed by silver staining. The protein specifically bound to GST-SSL5 (*) was subjected to peptide mass fingerprinting analysis.
To examine the role of glycans of C1 inhibitor in the binding to SSL5, we treated C1Inh-His with PNGase F or neuraminidase and subjected the resulting samples to a pull-down assay. We first analyzed these carbohydrate-modified samples by western blotting using anti-His-tag antibody. PNGase F-treated C1Inh-His migrated faster than the intact C1Inh-His in SDS-PAGE (Fig. 4A), probably due to the removal of N-glycans, whereas neuraminidase treatment did not significantly affect its mobility. Next, we analyzed the same samples by lectin blotting with a sialic acid-specific lectin, Maackia amurensis mitogen (MAM). Both PNGase F-treated and neuraminidase-treated C1Inh-His appeared to show decreased affinity for the lectin as compared with the intact C1Inh-His (Fig. 4A). Pull-down assay of the carbohydrate-modified C1Inh-His clearly indicated that GST-SSL5-agarose failed to bind to neuraminidasetreated C1Inh-His but had an affinity for PNGase F-treated C1InhHis similar to that of the intact C1Inh-His (Fig. 4B). These results suggest that the sialic acid residues of C1Inh-His are crucial for its interaction with GST-SSL5, but that N-glycans of C1Inh-His play only a limited role in the interaction. We then utilized Lec2 cells, which are deficient in the biosynthesis of sialic acid-containing glycans due to a mutation in the CMP-sialic acid transporter [24]. We purified C1Inh-His from the conditioned media of Lec2 cells and wild-type CHO-K1 cells
Please cite this article as: T. Oku et al., Identification of human plasma C1 inhibitor as a target protein for staphylococcal superantigen-like protein 5 (SSL5), Biochemical and Biophysical Research Communications, https://doi.org/10.1016/j.bbrc.2018.12.026
4
T. Oku et al. / Biochemical and Biophysical Research Communications xxx (xxxx) xxx
Fig. 3. Attenuation of the inhibitory activity of C1Inh-His against complement C1s by GST-SSL5. (A) Recombinant C1s (1 mg/mL) was incubated with the chromogenic substrate (N-carbobenzyloxy-Lys-thiobenzyl ester) (0.1 mM) and DTNB (0.1 mM) in the presence of recombinant C1Inh-His (0e100 mg/mL). The absorbance at 415 nm was measured with a microplate reader. (B) C1Inh-His (25 mg/mL) was incubated with GSTSSL5 (0e400 mg/mL) or GST (400 mg/mL) at room temperature for 30 min. Recombinant C1s was then added and incubated with the chromogenic substrate and DTNB as described above. The absorbance at 415 nm was measured, and data are presented as the mean ± SEM. ***p < 0.005 vs. control.
Fig. 2. Binding of GST-SSL5 to recombinant His-tagged C1 inhibitor as assessed by pull-down assay. (A) Schematic diagram of the structure of recombinant the Histagged C1 inhibitor with a flexible spacer (Gly4-Ser) at the C-terminus (C1Inh-His). (B) Purified C1Inh-His was separated by SDS-PAGE (5e20% gradient gel) and stained with CBB. (C) C1 esterase activity was measured in the presence of purified C1Inh-His (18 mg/mL) or human plasma (1.8%) by a functional chromogenic assay. Experiments were performed in triplicate, and the data are presented as the mean ± SEM. Statistical data analysis was conducted using the Student's t-test. ***p < 0.005 vs. control (PBS). (D) C1Inh-His was incubated with GST-SSL5-agarose or control GST-agarose, and the bound proteins were analyzed by SDS-PAGE (10%) and stained with CBB. An arrowhead indicates C1Inh-His. (E) GST-SSL5 or GST was incubated with C1Inh-His-loaded resin, and proteins bound to the resin were analyzed by SDS-PAGE (10%) and stained with CBB. An arrowhead indicates GST-SSL5.
transfected with C1Inh-His cDNA and analyzed the purified samples by SDS-PAGE (Fig. 4C). Both samples had almost the same mobility in SDS-PAGE. However, a pull-down assay using GST-SSL5agarose and western blotting using anti-His-tag antibody showed that immobilized SSL5 bound to C1Inh-His derived from the wildtype CHO-K1 cells but did not bind to C1Inh-His from Lec2 cells (Fig. 4D). 4. Discussion In the present study, we demonstrated that SSL5 specifically bound to human plasma C1Inh (Figs. 1 and 2), which is known as one of the key regulators in the classical pathway of complement activation [25]. The inhibitory activity of C1Inh toward C1s was reduced in the presence of recombinant GST-SSL5 (Fig. 3). Because
Please cite this article as: T. Oku et al., Identification of human plasma C1 inhibitor as a target protein for staphylococcal superantigen-like protein 5 (SSL5), Biochemical and Biophysical Research Communications, https://doi.org/10.1016/j.bbrc.2018.12.026
T. Oku et al. / Biochemical and Biophysical Research Communications xxx (xxxx) xxx
5
recombinant C1Inh with neuraminidase markedly decreased its affinity for SSL5 (Figs. 4B), and 2) recombinant C1Inh produced by Lec2 mutant cells had much lower affinity than that from the wildtype CHO-K1 cells (Fig. 4D). Human C1Inh is a heavily glycosylated protein consisting of ca. 49% carbohydrates: six N-linked and seven O-linked carbohydrate chains [32]. Because the binding affinity of C1Inh to SSL5 was apparently unchanged after the removal of Nglycans by PNGase F treatment (Fig. 4B), we conclude that sialic acid residues attached to N-glycans play a limited role in the C1Inh/ SSL5 interaction, but those attached to O-glycans might be important for the interaction. The possible involvement of O-glycans of C1Inh in the binding of C1Inh to SSL5 is consistent with our previous observation that sialic acid residues in O-glycans of human leukocyte MMP-9 played a crucial role in the binding of MMP-9 to SSL5 [9]. Recently, Stavenhagen and co-workers investigated the structures of O-glycans of human C1Inh in detail [33]. Their study indicated that human C1Inh carried mainly core 1-type O-glycans with sialylation and core 2-type O-glycans as minor components. They also detected a domain rich in O-glycans in the N-terminal domain of the molecule, and this domain might be important for C1Inh to interact with SSL5. One of the common structural characteristics among several proteins so far identified as SSL5-binding proteinsdi.e., PSGL-1, platelet membranes GPIb and GPVI, leukocyte MMP-9 and plasma C1Inhdis the presence of heavily O-glycosylated mucin-like domains [6e9], and SSL5 is likely to recognize such domains with clustered O-glycans in these glycoproteins. This study showed a novel effect of SSL5 on a complement regulator. In addition to the previous observation that SSL5 had suppressive effects on leukocyte trafficking through the inhibition of the leukocyte/endothelium interaction and MMP-9 enzymatic activity, SSL5 may also modulate complement activation. In addition, the previous studies, including ours, showed that two other members of the SSL family of proteins, SSL7 and SSL10, bound to immunoglobulins and complement components, resulting in the impairment of complement functions. These exoproteins may cooperatively induce perturbation of the host immune system and facilitate S. aureus infection. Fig. 4. Effect of modification of glycans of C1Inh-His on its binding to GST-SSL5. (A) C1Inh-His was treated with PNGase F (PNG) or neuraminidase (Neu) at 37 C for 2 h and analyzed by western blotting using anti-His-tag antibody (left panel) or a lectin from Maackia amurensis (MAM) (right panel). (B) C1Inh-His treated with PNGase F or neuraminidase was subjected to pull-down assay using GST-SSL5-agarose. The bound proteins were analyzed by western blotting using anti-His tag antibody (upper panel) and anti-GST antibody (lower panel). (C) C1Inh-His was purified from the conditioned media of Lec2 mutant or wild-type CHO-K1 cells after cDNA transfection and analyzed by SDS-PAGE (7.5%) followed by western blotting using anti-His tag antibody. (D) C1Inh-His purified from the conditioned media of Lec2 cells or CHO-K1 cells was subjected to pull-down assay using GST-SSL5-agarose. The bound proteins were analyzed by western blotting using anti-His-tag antibody (upper panel) and anti-GST antibody (lower panel).
Authors’ roles TO conceived the project and designed the experiments. TO, CK and YA performed the experiments. TO, CK and TT wrote and edited the manuscript. TT supervised the research. Conflicts of interest There are no conflicts of interest to declare. Ethical approval and consent to participate
C1Inh plays a role in maintenance of the classical pathway by preventing excessive complement activation, we consider that abrogation of C1Inh by SSL5 may disturb the host immune reactions. C1Inh is a multifunctional serine protease inhibitor and is also known to regulate MASP proteases of the lectin pathway of complement activation [26e28], the blood coagulation cascade [29] and the kallikrein/kinin system [30,31]. In patients with hereditary angioedema (HAE), a deficiency or dysfunction of C1Inh causes inappropriate activation of the kallikrein/kinin system. It has not been clarified whether SSL5 abrogates the suppression by C1Inh toward all these proteases, but SSL5 might have a major influence on a wide range of biological defense mechanisms. In the present experiments, the interaction between C1Inh and SSL5 was suggested to be mediated by sialic acid-containing glycans of C1Inh based on the following results: 1) the treatment of
We did not use samples collected from patients or animals in any of the experiments. Consent for publication All the authors have approved the manuscript and agree with its submission. Acknowledgments We are grateful to Dr. Misao Matsushita (Department of Applied Biochemistry, Tokai University) for his helpful discussion. We also thank Ms. Ayako Shimo, Ms. Keiko Otani, Ms. Yukari Yakawa, and Ms. Airi Watanabe (Hoshi University School of Pharmacy and Pharmaceutical Sciences) for their technical assistance. This work
Please cite this article as: T. Oku et al., Identification of human plasma C1 inhibitor as a target protein for staphylococcal superantigen-like protein 5 (SSL5), Biochemical and Biophysical Research Communications, https://doi.org/10.1016/j.bbrc.2018.12.026
6
T. Oku et al. / Biochemical and Biophysical Research Communications xxx (xxxx) xxx
was supported in part by Grants-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology of Japan and by Research Grants from the Shiono Wellness Foundation and the Japan Agency for Medical Research and Development (AMED). References [1] A.N. Spaan, B.G. Surewaard, R. Nijland, et al., Neutrophils versus Staphylococcus aureus: a biological tug of war, Annu. Rev. Microbiol. 67 (2013) 629e650. [2] T.J. Foster, Immune evasion by staphylococci, Nat. Rev. Microbiol. 3 (2005) 948e958. [3] V. Thammavongsa, H.K. Kim, D. Missiakas, et al., Staphylococcal manipulation of host immune responses, Nat. Rev. Microbiol. 13 (2015) 529e543. [4] J.D. Fraser, T. Proft, The bacterial superantigen and superantigen-like proteins, Immunol. Rev. 225 (2008) 226e243. [5] R.J. Williams, J.M. Ward, B. Henderson, et al., Identification of a novel gene cluster encoding staphylococcal exotoxin-like proteins: characterization of the prototypic gene and its protein product, SET1, Infect. Immun. 68 (2000) 4407e4415. [6] J. Bestebroer, M.J. Poppelier, L.H. Ulfman, et al., Staphylococcal superantigenlike 5 binds PSGL-1 and inhibits P-selectin-mediated neutrophil rolling, Blood 109 (2007) 2936e2943. [7] C.J. de Haas, C. Weeterings, M.M. Vughs, et al., Staphylococcal superantigenlike 5 activates platelets and supports platelet adhesion under flow conditions, which involves glycoprotein Ibalpha and alpha IIb beta 3, J. Thromb. Haemostasis 7 (2009) 1867e1874. [8] S. Itoh, E. Hamada, G. Kamoshida, et al., Staphylococcal superantigen-like protein 5 inhibits matrix metalloproteinase 9 from human neutrophils, Infect. Immun. 78 (2010) 3298e3305. [9] C. Kurisaka, T. Oku, S. Itoh, et al., Role of sialic acid-containing glycans of matrix metalloproteinase-9 (MMP-9) in the interaction between MMP-9 and staphylococcal superantigen-like protein 5, Microbiol. Immunol. 62 (2018) 168e175. [10] I. Jongerius, J. Kohl, M.K. Pandey, et al., Staphylococcal complement evasion by various convertase-blocking molecules, J. Exp. Med. 204 (2007) 2461e2471. [11] H. Chen, D. Ricklin, M. Hammel, et al., Allosteric inhibition of complement function by a staphylococcal immune evasion protein, Proc. Natl. Acad. Sci. U. S. A. 107 (2010) 17621e17626. [12] A.J. Laarman, M. Ruyken, C.L. Malone, et al., Staphylococcus aureus metalloprotease aureolysin cleaves complement C3 to mediate immune evasion, J. Immunol. 186 (2011) 6445e6453. [13] R. Langley, B. Wines, N. Willoughby, et al., The staphylococcal superantigenlike protein 7 binds IgA and complement C5 and inhibits IgA-Fc alpha RI binding and serum killing of bacteria, J. Immunol. 174 (2005) 2926e2933. [14] P.A. Ramsland, N. Willoughby, H.M. Trist, et al., Structural basis for evasion of IgA immunity by Staphylococcus aureus revealed in the complex of SSL7 with Fc of human IgA1, Proc. Natl. Acad. Sci. U. S. A. 104 (2007) 15051e15056. [15] N.S. Laursen, N. Gordon, S. Hermans, et al., Structural basis for inhibition of complement C5 by the SSL7 protein from Staphylococcus aureus, Proc. Natl. Acad. Sci. U. S. A. 107 (2010) 3681e3686.
[16] S. Itoh, E. Hamada, G. Kamoshida, et al., Staphylococcal superantigen-like protein 10 (SSL10) binds to human immunoglobulin G (IgG) and inhibits complement activation via the classical pathway, Mol. Immunol. 47 (2010) 932e938. [17] S. Itoh, R. Yokoyama, G. Kamoshida, et al., Staphylococcal superantigen-like protein 10 (SSL10) inhibits blood coagulation by binding to prothrombin and factor Xa via their gamma-carboxyglutamic acid (Gla) domain, J. Biol. Chem. 288 (2013) 21569e21580. [18] D. Patel, B.D. Wines, R.J. Langley, et al., Specificity of staphylococcal superantigen-like protein 10 toward the human IgG1 Fc domain, J. Immunol. 184 (2010) 6283e6292. [19] T. Oku, H. Soma, C. Kurisaka, et al., Generation of a monoclonal antibody against staphylococcal superantigen-like protein 5 (SSL5) that discriminates SSL5 from other SSL proteins, Monoclon. Antibodies Immunodiagn. Immunother. 37 (2018) 212e217. [20] T. Oku, S. Itoh, M. Okano, et al., Two regions responsible for the actin binding of p57, a mammalian coronin family actin-binding protein, Biol. Pharm. Bull. 26 (2003) 409e416. [21] T. Oku, S. Itoh, R. Ishii, et al., Homotypic dimerization of the actin-binding protein p57/coronin-1 mediated by a leucine zipper motif in the C-terminal region, Biochem. J. 387 (2005) 325e331. [22] X. Chen, J.L. Zaro, W.C. Shen, Fusion protein linkers: property, design and functionality, Adv. Drug Deliv. Rev. 65 (2013) 1357e1369. [23] T. Oku, Y. Ando, M. Ogura, et al., Development of splice variant-specific monoclonal antibodies against human alpha3 integrin, Monoclon. Antibodies Immunodiagn. Immunother. 35 (2016) 12e17. [24] M. Eckhardt, B. Gotza, R. Gerardy-Schahn, Mutants of the CMP-sialic acid transporter causing the Lec2 phenotype, J. Biol. Chem. 273 (1998) 20189e20195. [25] A.E. Davis III, P. Mejia, F. Lu, Biological activities of C1 inhibitor, Mol. Immunol. 45 (2008) 4057e4063. [26] M. Matsushita, S. Thiel, J.C. Jensenius, et al., Proteolytic activities of two types of mannose-binding lectin-associated serine protease, J. Immunol. 165 (2000) 2637e2642. [27] J.S. Presanis, K. Hajela, G. Ambrus, et al., Differential substrate and inhibitor profiles for human MASP-1 and MASP-2, Mol. Immunol. 40 (2004) 921e929. [28] L. Beinrohr, J. Dobo, P. Zavodszky, et al., MBL-MASPs and C1-inhibitor: novel approaches for targeting complement-mediated inflammation, Trends Mol. Med. 14 (2008) 511e521. [29] M. Cugno, I. Bos, Y. Lubbers, et al., In vitro interaction of C1-inhibitor with thrombin, Blood Coagul. Fibrinolysis 12 (2001) 253e260. [30] I. Gigli, J.W. Mason, R.W. Colman, et al., Interaction of plasma kallikrein with the C1 inhibitor, J. Immunol. 104 (1970) 574e581. [31] P.C. Harpel, M.F. Lewin, A.P. Kaplan, Distribution of plasma kallikrein between C-1 inactivator and alpha 2-macroglobulin in plasma utilizing a new assay for alpha 2-macroglobulin-kallikrein complexes, J. Biol. Chem. 260 (1985) 4257e4263. [32] S.C. Bock, K. Skriver, E. Nielsen, et al., Human C1 inhibitor: primary structure, cDNA cloning, and chromosomal localization, Biochemistry 25 (1986) 4292e4301. [33] K. Stavenhagen, H.M. Kayili, S. Holst, et al., N- and O-glycosylation analysis of human C1-inhibitor reveals extensive mucin-type O-glycosylation, Mol. Cell. Proteomics 17 (2018) 1225e1238.
Please cite this article as: T. Oku et al., Identification of human plasma C1 inhibitor as a target protein for staphylococcal superantigen-like protein 5 (SSL5), Biochemical and Biophysical Research Communications, https://doi.org/10.1016/j.bbrc.2018.12.026
Contents lists available at ScienceDirect
Biochemical and Biophysical Research Communications journal homepage: www.elsevier.com/locate/ybbrc
Identification of human plasma C1 inhibitor as a target protein for staphylococcal superantigen-like protein 5 (SSL5) Teruaki Oku*, Chisato Kurisaka, Yusuke Ando, Tsutomu Tsuji** Department of Microbiology, Hoshi University School of Pharmacy and Pharmaceutical Sciences, Tokyo, Japan
a r t i c l e i n f o
a b s t r a c t
Article history: Received 27 November 2018 Accepted 5 December 2018 Available online xxx
The family of staphylococcal superantigen-like proteins (SSLs) have a structure similar to bacterial superantigens but exhibit no superantigenic activity. These exoproteins have recently been shown to disturb the host immune defense system. One family member, SSL5, was reported to bind to human leukocyte P-selectin glycoprotein ligand-1 (PSGL-1) and matrix metalloproteinase-9 (MMP-9) and to interfere with leukocyte trafficking. In the present study, we explored human plasma proteins bound by glutathione S-transferase (GST)-tagged recombinant SSL5 (GST-SSL5) and identified plasma protease C1 inhibitor (C1Inh) as a major SSL5-binding protein based on the results of peptide mass fingerprinting analysis with MALDI-TOFMS. GST-SSL5 was found to attenuate the inhibitory activity of recombinant histidine-tagged C1Inh (C1Inh-His) toward complement C1s. We also observed that the treatment of C1Inh-His with neuraminidase markedly decreased its binding to GST-SSL5. Moreover, C1Inh-His produced by Lec2 mutant cells (deficient in sialic acid biosynthesis) showed much lower binding affinity for SSL5 than that produced by the wild-type CHO-K1 cells, as assessed by pull-down assay. These results suggest that SSL5 binds to C1Inh in a sialic acid-dependent fashion and modulates the host immune defense through perturbation of the complement system in association with S. aureus infection. © 2018 Elsevier Inc. All rights reserved.
Keywords: Staphylococcus aureus SSL5 C1 inhibitor Complement Immune perturbation Sialic acid
1. Introduction Staphylococcus aureus is Gram-positive bacterium, and its infection is considered to be the most frequent cause of a variety of diseases ranging from mild skin infection to serious infection such as sepsis [1]. S. aureus secretes various proteins that bind to critical molecules involved in the host innate and adaptive immune system as well as cytotoxins [2,3]. Staphylococcal superantigen-like proteins (SSLs) are a family of exoproteins that have a structure similar to bacterial superantigens such as toxic shock syndrome toxine-1 (TSST-1) and streptococcal pyrogenic exotoxin C (SPEC) but have no superantigenic activity because they lack a T cell receptor (TCR)-binding domain [4]. The family is composed of 14 members
Abbreviations: SSL, staphylococcal superantigen-like protein; GST, glutathione S-transferase; C1Inh, plasma protease C1 inhibitor; MMP-9, matrix metalloproteinase-9; PBS, phosphate-buffered saline; SDS-PAGE, sodium dodecyl sulfatepolyacrylamide gel electrophoresis; MALDI-TOFMS, matrix-assisted laser desorption/ionization-time of flight mass spectrometry; HRP, horseradish peroxidase; DTNB 5, 5’-dithiobis(2-nitrobenzoic acid); PNGase F, peptide-N-glycosidase F. * Corresponding author. ** Corresponding author. E-mail addresses: [email protected] (T. Oku), [email protected] (T. Tsuji).
(SSL1-SSL14), and these proteins possess a common structure, the N-terminal b-barrel globular domain (OB-fold) and C-terminal bgrasp domain [4,5]. Previous studies have indicated that SSL5 binds specifically to P-selectin glycoprotein ligand-1 (PSGL-1) on leukocyte cell membranes [6] and to platelet membrane glycoproteins (GPIb and GPVI) [7]. We also reported that SSL5 bound to matrix metalloproteinase-9 (MMP-9) from human neutrophils and inhibited its enzymatic activity [8]. These results strongly suggest that SSL5 has significant influences on leukocyte trafficking and platelet activation. Because PSGL-1, GPIb, GPVI and MMP-9 are glycoproteins with rather high carbohydrate contents, the carbohydrate chains of these glycoproteins are thought to be important for their interactions with SSL5. Indeed, we found that the interaction between SSL5 and MMP-9 involved sialylated O-glycans of MMP-9 [9]. Coagulase produced by S. aureus is a well-known exoenzyme that binds to plasma prothrombin to form a complex with catalytic activity for the conversion of fibrinogen to fibrin. In addition, it has been reported that S. aureus secretes several exoproteins reacting with plasma proteins, including blood coagulation factors and complement components [10e12]. Some SSL proteins were also shown to bind to plasma proteins; e.g., SSL7 bound to human IgA
https://doi.org/10.1016/j.bbrc.2018.12.026 0006-291X/© 2018 Elsevier Inc. All rights reserved.
Please cite this article as: T. Oku et al., Identification of human plasma C1 inhibitor as a target protein for staphylococcal superantigen-like protein 5 (SSL5), Biochemical and Biophysical Research Communications, https://doi.org/10.1016/j.bbrc.2018.12.026
2
T. Oku et al. / Biochemical and Biophysical Research Communications xxx (xxxx) xxx
and complement C5 [13e15], and SSL10 bound to human IgG, prothrombin and factor Xa [16e18]. Because SSLs suppressed phagocytosis by leukocytes, blood coagulation and complement activation through these interactions, SSLs were considered to play roles in the immune evasion of S. aureus. In this study, therefore, we attempted to explore possible target plasma proteins for SSL5 by using affinity purification, and found that recombinant SSL5 specifically bound to a complement regulator. 2. Materials and methods 2.1. Reagents Isopropyl-b-D-thiogalactopyranoside (IPTG), glutathione, and imidazole were obtained from FUJIFILM Wako Pure Chemical Corp. (Osaka, Japan). Brij 35, RPMI1640 medium, trypsin, 5, 5’-dithiobis(2-nitrobenzoic acid) (DTNB), Triton X-100, and TriReagent were purchased from Sigma-Aldrich (St. Louis, MO, USA). One step Coomassie Brilliant Blue (CBB) was purchased from Bio Craft (Tokyo, Japan). PrimeSTAR MAX DNA polymerase was obtained from Takara Bio Inc. (Shiga, Japan). Oligonucleotides were supplied by FASMAC (Kanagawa, Japan). cOmplete™ His-Tag Purification Resin and 1,4-dithiothreitol (DTT) were purchased from Roche Diagnostics (Indianapolis, IN, USA). Glutathione-agarose was obtained from Thermo Fisher Scientific (Waltham, MA, USA). Human pooled plasma and fetal calf serum (FCS) were supplied by Kohjin Bio (Saitama, Japan) and Biosera (Boussens, France), respectively. A monoclonal antibody against glutathione S-transferase (GST) was prepared in our laboratory [19]. Anti-6 His mouse monoclonal antibody and horseradish peroxidase (HRP)-conjugated goat antimouse IgG antibody were purchased from Thermo Fisher Scientific and Kirkegaard & Perry Laboratories (Gaithersburg, MD, USA), respectively. Biotin-conjugated Maackia amurensis mitogen (MAM) was purchased from Cosmo Bio Co., Ltd. (Tokyo, Japan). HRPconjugated streptavidin was obtained from BioLegend (San Diego, CA, USA). Peptide-N-glycosidase F (PNGase F) and neuraminidase (Salmonella typhimurium LT2) were purchased from New England Biolabs (Ipswich, MA, USA).
0.25% Brij 35 at 4 C for 1 h. After the beads were washed with the same buffer, proteins bound to the beads were recovered with SDS sample buffer (50 mM Tris-HCl, 1% SDS, 5% glycerol, 0.01% bromophenol blue, pH 6.8) and analyzed by SDS-PAGE and silver staining. 2.4. In-gel digestion and peptide mass fingerprinting (PMF) The protein band was excised from the SDS-PAGE gel stained with silver nitrate, destained with 15 mM potassium ferricyanide and 50 mM sodium thiosulfate, reduced with 10 mM DTT in 25 mM ammonium bicarbonate, and alkylated with 55 mM iodoacetamide in 25 mM ammonium bicarbonate in a water/acetonitrile (50/50) mixture. The reduced/alkylated protein was digested by treatment of the gel with trypsin (10 mg/mL) in 45 mM ammonium bicarbonate. The digested peptides were extracted with 5% trifluoroacetic acid in 50% acetonitrile. Samples were concentrated under reduced pressure and purified with C18 ZipTips (Millipore, Billerica, MA, USA) following the manufacturer's instructions. Peptide mass fingerprinting analysis was performed using MALDI-TOFMS (AXIMA-QIT; Shimadzu, Kyoto, Japan), and monoisotopic masses were processed for identification through analysis with m/z software and a search of the SwissProt database using MASCOT software (Matrix Science, Boston, MA, USA). 2.5. Preparation of recombinant C1 inhibitor (C1Inh-His)
The SSL5 gene was inserted into pGEX-5X-1 vector (GE Healthcare, Piscataway, NJ, USA) as described previously [9]. The recombinant GST-tagged SSL5 (GST-SSL5) was expressed in Escherichia coli (SoluBL21; Genlantis, San Diego, CA, USA), and purified as described previously [20,21]. Briefly, the transformed E. coli grown to exponential phase was cultured further in TB medium (12 g/L tryptone, 24 g/L yeast extract, 0.4% (v/v) glycerol, 17 mM KH2PO4 and 72 mM K2HPO4) containing 0.2 mM IPTG at 25 C for 16 h. The cells harvested by centrifugation were resuspended in sonication buffer (50 mM Tris/HCl, pH 8.0, containing 150 mM NaCl and 1 mM EDTA), and lysozyme (final 1.0 mg/mL) was added to the suspension. The cells were disrupted by sonication and centrifuged. The supernatant was mixed with glutathione agarose (Thermo Fisher Scientific) and incubated at 4 C for 1 h with gentle agitation. After washing with sonication buffer, GST fusion protein was eluted with 10 mM of glutathione (reduced form). Purified protein was applied to a PD-10 column (GE Healthcare) to replace the buffer with phosphate-buffered saline (PBS).
A DNA construct for the expression of a fusion protein of human plasma protease C1 inhibitor with His-tag (C1Inh-His) with a flexible spacer (Gly4-Ser) [22] at the C-terminus of C1Inh (C1IN) was amplified by PCR using cDNA derived from the human mononuclear leukocytes as a template. PCR was conducted with the following set of primers: 50 -AGC GGC CGC CAC CAT GGC CTC CAG GCT G-3’ (sense) and 50 -CTC GAG TCA ATG GTG ATG GTG ATG ATG GGA TCC ACC ACC TCC GGC CCT GGG GTC ATA TAC TCG-3’ (antisense). The product was cloned into the NotI/XhoI restriction site of pcDNA6-V5HisA (Thermo Fisher Scientific) (pcDNA6-C1Inh-His). The full-length cDNA sequence of the C1Inh-His was determined by the dideoxy method. The plasmid (pcDNA6-C1Inh-His) was introduced into HEK293FT (Thermo Fisher Scientific), CHO-K1 or Lec2 cells (CRL1736; American Type Culture Collection) with polyethylenimine MAX (PEI) (Mw 40,000) as described previously [23]. The culture supernatant (500 mL) was mixed with cOmplete His-tag purification resin (Roche Diagnostics) and incubated 1 h at 4 C with gentle agitation. The resin was washed with 50 mM Tris-HCl, 500 mM NaCl, and 15 mM imidazole (pH 7.4), and the C1Inh-His bound to the resin was eluted with 50 mM Tris-HCl, 500 mM NaCl, and 200 mM imidazole (pH 7.4). Purified C1Inh-His was applied to a PD10 column to replace the buffer with PBS. Inhibitory activity of C1Inh-His against C1 esterase was measured by a functional chromogenic assay (SIEMENS, Eschborn, Germany) according to the manufacturer's instructions. Briefly, purified C1Inh-His (2 mL, 1 mg/mL) or human plasma was treated with C1 esterase (0.1 mL) at 37 C for 5 min. The mixture was incubated with a synthetic substrate for C1 esterase (MeOC-Lys(εCbo)Gly-Arg-pNA, 5 mM) (0.01 mL) at 37 C for 5 min. The reaction was stopped with 20% acetic acid (0.05 mL), and absorbance at 415 nm was measured with a microplate reader (MTP-450; Corona, Tokyo, Japan).
2.3. Pull-down assay
2.6. Measurement of C1s activity
Purified GST-SSL5 or GST (5 mg) was bound to glutathione agarose (30 mL, 50% slurry) at 4 C for 1 h. The agarose beads were incubated with human plasma (100 mL) diluted with PBS containing
Complement C1s activity was measured using a chromogenic substrate, N-carbobenzyloxy-Lys-thiobenzyl ester (Bachem, Bubendorf, Switzerland), and DTNB. The assay was performed
2.2. Preparation of recombinant GST-SSL5
Please cite this article as: T. Oku et al., Identification of human plasma C1 inhibitor as a target protein for staphylococcal superantigen-like protein 5 (SSL5), Biochemical and Biophysical Research Communications, https://doi.org/10.1016/j.bbrc.2018.12.026
T. Oku et al. / Biochemical and Biophysical Research Communications xxx (xxxx) xxx
according to the manufacturer's instructions using recombinant complement C1s (Merck, Bedford, MA, USA) as a positive control. Experiments were performed in triplicate, and statistical data analysis was conducted using the Student's t-test. 2.7. Treatments with PNGase F and neuraminidase N-glycans of purified C1Inh-His (1 mg) were removed by incubation with PNGase F (500 units) in 50 mM sodium phosphate buffer (pH 6.0) at 37 C for 3 h. Sialic acid residues were hydrolyzed with neuraminidase (50 units) in 50 mM sodium acetate buffer (pH 5.5) containing 5 mM CaCl2 at 37 C for 3 h. 3. Results 3.1. Recombinant GST-SSL5 binds to C1 inhibitor from human plasma The purified recombinant GST-SSL5 and GST were analyzed by SDS-PAGE, and each protein gave an apparently single band at ~48 kDa (GST-SSL5) or ~25 kDa (GST) after staining with CBB (Fig. 1A). We attempted to isolate SSL5-binding proteins from human plasma by pull-down assay. When the proteins bound by GSTSSL5 were analyzed by SDS-PAGE and silver staining, a major protein band of ~100 kDa was detected (Fig. 1B). This protein was analyzed by peptide mass fingerprinting with MALDI-TOFMS and identified as plasma protease C1 inhibitor (C1Inh; also called C1 inhibitor, C1 esterase inhibitor and complement C1 inactivator) by the MASCOT search engine. The mass peaks of amino acid sequences corresponding to human C1 inhibitor (UniProt: P05155) detected by MS were as follows: m/z 853.45 for 488FPVFMGR494, m/z 910.48 for 242TLYSSSPR249, m/z 935.46 for 310MEPFHFK316, m/z 1067.53 for 153LYHAFSAMK161, m/z 1116.59 for 277LLDSLPSDTR286, 391 m/z 1185.70 for FQPTLLTLPR400, m/z 1218.59 for 202 211 DFTCVHQALK , m/z 1264.68 for 191TNLESILSYPK201, m/z 1317.80 for 287LVLLNAIYLSAK298, m/z 1431.73 for 330YPVAHFIDQTLK341, m/z 1482.75 for 403VTTSQDMLSIMEK415, m/z 1559.87 for 329KYPVAHFIDQTLK341, m/z 1593.79 for 367LEDMEQALSPSVFK380, m/z 1826.98
217 for GVTSVSQIFHSPDLAIR233 HRLEDMEQALSPSVFK380.
3
and
m/z
1886.97
for
365
3.2. SSL5 binds to recombinant C1 inhibitor To further examine the interaction between SSL5 and C1 inhibitor, we prepared a recombinant histidine-tagged C1 inhibitor (C1Inh-His) (Fig. 2A). C1Inh-His expressed in HEK293FT cells was purified from the culture supernatants with His-tag purification resin. SDS-PAGE analysis of the purified protein gave an apparently single band at ~100 kDa after staining with CBB (Fig. 2B). The inhibitory activity of C1Inh-His against C1 esterase was measured using a synthetic substrate. The addition of purified C1Inh-His (18 mg/mL) decreased the esterase activity to approximately 45%, a level comparable to that in human plasma (1.8%) (Fig. 2C). Next, we performed a pull-down assay using immobilized GST-SSL5 and soluble C1Inh-His followed by SDS-PAGE and CBB staining. As shown in Fig. 2D, the fraction bound to GST-SSL5-agarose gave the protein band corresponding to C1Inh-His (~100 kDa), whereas no band was detected from the fraction bound to GST-agarose. Conversely, the pull-down assay using immobilized C1Inh-His and soluble GST-SSL5 resulted in the binding of GST-SSL5 but not GST to immobilized C1Inh-His (Fig. 2E). 3.3. SSL5 attenuates C1 inhibitor activity C1 inhibitor shows inhibitory activity against complement components C1s and C1r and thus regulates complement activation. We also confirmed that C1Inh-His inhibited C1s activity in a dose-dependent manner as measured with a colorimetric peptide substrate, N-carbobenzyloxy-Lys-thiobenzyl ester (Z-K-SBzl) (Fig. 3A). We next examined the effect of GST-SSL5 on the inhibitory activity of C1Inh-His toward C1s. As shown in Fig. 3B, C1s activity inhibited in the presence of C1Inh-His was partially recovered by the preincubation with GST-SSL5 but not with control GST; i.e., the 66% inhibition observed in the presence of C1Inh-His (25 mg/mL) was reduced to 28% inhibition after the addition of GST-SSL5 (400 mg/mL). 3.4. Sialic acid-containing glycans of C1 inhibitor are involved in the interaction with SSL5
Fig. 1. Isolation of SSL5-binding proteins from human plasma. (A) Purified recombinant GST-SSL5 and the control GST were separated by SDS-PAGE (5e20% gradient gel) and stained with CBB. (B) Human plasma was incubated with GST-SSL5agarose or control GST-agarose, and the bound proteins were separated by SDS-PAGE (7.5%) followed by silver staining. The protein specifically bound to GST-SSL5 (*) was subjected to peptide mass fingerprinting analysis.
To examine the role of glycans of C1 inhibitor in the binding to SSL5, we treated C1Inh-His with PNGase F or neuraminidase and subjected the resulting samples to a pull-down assay. We first analyzed these carbohydrate-modified samples by western blotting using anti-His-tag antibody. PNGase F-treated C1Inh-His migrated faster than the intact C1Inh-His in SDS-PAGE (Fig. 4A), probably due to the removal of N-glycans, whereas neuraminidase treatment did not significantly affect its mobility. Next, we analyzed the same samples by lectin blotting with a sialic acid-specific lectin, Maackia amurensis mitogen (MAM). Both PNGase F-treated and neuraminidase-treated C1Inh-His appeared to show decreased affinity for the lectin as compared with the intact C1Inh-His (Fig. 4A). Pull-down assay of the carbohydrate-modified C1Inh-His clearly indicated that GST-SSL5-agarose failed to bind to neuraminidasetreated C1Inh-His but had an affinity for PNGase F-treated C1InhHis similar to that of the intact C1Inh-His (Fig. 4B). These results suggest that the sialic acid residues of C1Inh-His are crucial for its interaction with GST-SSL5, but that N-glycans of C1Inh-His play only a limited role in the interaction. We then utilized Lec2 cells, which are deficient in the biosynthesis of sialic acid-containing glycans due to a mutation in the CMP-sialic acid transporter [24]. We purified C1Inh-His from the conditioned media of Lec2 cells and wild-type CHO-K1 cells
Please cite this article as: T. Oku et al., Identification of human plasma C1 inhibitor as a target protein for staphylococcal superantigen-like protein 5 (SSL5), Biochemical and Biophysical Research Communications, https://doi.org/10.1016/j.bbrc.2018.12.026
4
T. Oku et al. / Biochemical and Biophysical Research Communications xxx (xxxx) xxx
Fig. 3. Attenuation of the inhibitory activity of C1Inh-His against complement C1s by GST-SSL5. (A) Recombinant C1s (1 mg/mL) was incubated with the chromogenic substrate (N-carbobenzyloxy-Lys-thiobenzyl ester) (0.1 mM) and DTNB (0.1 mM) in the presence of recombinant C1Inh-His (0e100 mg/mL). The absorbance at 415 nm was measured with a microplate reader. (B) C1Inh-His (25 mg/mL) was incubated with GSTSSL5 (0e400 mg/mL) or GST (400 mg/mL) at room temperature for 30 min. Recombinant C1s was then added and incubated with the chromogenic substrate and DTNB as described above. The absorbance at 415 nm was measured, and data are presented as the mean ± SEM. ***p < 0.005 vs. control.
Fig. 2. Binding of GST-SSL5 to recombinant His-tagged C1 inhibitor as assessed by pull-down assay. (A) Schematic diagram of the structure of recombinant the Histagged C1 inhibitor with a flexible spacer (Gly4-Ser) at the C-terminus (C1Inh-His). (B) Purified C1Inh-His was separated by SDS-PAGE (5e20% gradient gel) and stained with CBB. (C) C1 esterase activity was measured in the presence of purified C1Inh-His (18 mg/mL) or human plasma (1.8%) by a functional chromogenic assay. Experiments were performed in triplicate, and the data are presented as the mean ± SEM. Statistical data analysis was conducted using the Student's t-test. ***p < 0.005 vs. control (PBS). (D) C1Inh-His was incubated with GST-SSL5-agarose or control GST-agarose, and the bound proteins were analyzed by SDS-PAGE (10%) and stained with CBB. An arrowhead indicates C1Inh-His. (E) GST-SSL5 or GST was incubated with C1Inh-His-loaded resin, and proteins bound to the resin were analyzed by SDS-PAGE (10%) and stained with CBB. An arrowhead indicates GST-SSL5.
transfected with C1Inh-His cDNA and analyzed the purified samples by SDS-PAGE (Fig. 4C). Both samples had almost the same mobility in SDS-PAGE. However, a pull-down assay using GST-SSL5agarose and western blotting using anti-His-tag antibody showed that immobilized SSL5 bound to C1Inh-His derived from the wildtype CHO-K1 cells but did not bind to C1Inh-His from Lec2 cells (Fig. 4D). 4. Discussion In the present study, we demonstrated that SSL5 specifically bound to human plasma C1Inh (Figs. 1 and 2), which is known as one of the key regulators in the classical pathway of complement activation [25]. The inhibitory activity of C1Inh toward C1s was reduced in the presence of recombinant GST-SSL5 (Fig. 3). Because
Please cite this article as: T. Oku et al., Identification of human plasma C1 inhibitor as a target protein for staphylococcal superantigen-like protein 5 (SSL5), Biochemical and Biophysical Research Communications, https://doi.org/10.1016/j.bbrc.2018.12.026
T. Oku et al. / Biochemical and Biophysical Research Communications xxx (xxxx) xxx
5
recombinant C1Inh with neuraminidase markedly decreased its affinity for SSL5 (Figs. 4B), and 2) recombinant C1Inh produced by Lec2 mutant cells had much lower affinity than that from the wildtype CHO-K1 cells (Fig. 4D). Human C1Inh is a heavily glycosylated protein consisting of ca. 49% carbohydrates: six N-linked and seven O-linked carbohydrate chains [32]. Because the binding affinity of C1Inh to SSL5 was apparently unchanged after the removal of Nglycans by PNGase F treatment (Fig. 4B), we conclude that sialic acid residues attached to N-glycans play a limited role in the C1Inh/ SSL5 interaction, but those attached to O-glycans might be important for the interaction. The possible involvement of O-glycans of C1Inh in the binding of C1Inh to SSL5 is consistent with our previous observation that sialic acid residues in O-glycans of human leukocyte MMP-9 played a crucial role in the binding of MMP-9 to SSL5 [9]. Recently, Stavenhagen and co-workers investigated the structures of O-glycans of human C1Inh in detail [33]. Their study indicated that human C1Inh carried mainly core 1-type O-glycans with sialylation and core 2-type O-glycans as minor components. They also detected a domain rich in O-glycans in the N-terminal domain of the molecule, and this domain might be important for C1Inh to interact with SSL5. One of the common structural characteristics among several proteins so far identified as SSL5-binding proteinsdi.e., PSGL-1, platelet membranes GPIb and GPVI, leukocyte MMP-9 and plasma C1Inhdis the presence of heavily O-glycosylated mucin-like domains [6e9], and SSL5 is likely to recognize such domains with clustered O-glycans in these glycoproteins. This study showed a novel effect of SSL5 on a complement regulator. In addition to the previous observation that SSL5 had suppressive effects on leukocyte trafficking through the inhibition of the leukocyte/endothelium interaction and MMP-9 enzymatic activity, SSL5 may also modulate complement activation. In addition, the previous studies, including ours, showed that two other members of the SSL family of proteins, SSL7 and SSL10, bound to immunoglobulins and complement components, resulting in the impairment of complement functions. These exoproteins may cooperatively induce perturbation of the host immune system and facilitate S. aureus infection. Fig. 4. Effect of modification of glycans of C1Inh-His on its binding to GST-SSL5. (A) C1Inh-His was treated with PNGase F (PNG) or neuraminidase (Neu) at 37 C for 2 h and analyzed by western blotting using anti-His-tag antibody (left panel) or a lectin from Maackia amurensis (MAM) (right panel). (B) C1Inh-His treated with PNGase F or neuraminidase was subjected to pull-down assay using GST-SSL5-agarose. The bound proteins were analyzed by western blotting using anti-His tag antibody (upper panel) and anti-GST antibody (lower panel). (C) C1Inh-His was purified from the conditioned media of Lec2 mutant or wild-type CHO-K1 cells after cDNA transfection and analyzed by SDS-PAGE (7.5%) followed by western blotting using anti-His tag antibody. (D) C1Inh-His purified from the conditioned media of Lec2 cells or CHO-K1 cells was subjected to pull-down assay using GST-SSL5-agarose. The bound proteins were analyzed by western blotting using anti-His-tag antibody (upper panel) and anti-GST antibody (lower panel).
Authors’ roles TO conceived the project and designed the experiments. TO, CK and YA performed the experiments. TO, CK and TT wrote and edited the manuscript. TT supervised the research. Conflicts of interest There are no conflicts of interest to declare. Ethical approval and consent to participate
C1Inh plays a role in maintenance of the classical pathway by preventing excessive complement activation, we consider that abrogation of C1Inh by SSL5 may disturb the host immune reactions. C1Inh is a multifunctional serine protease inhibitor and is also known to regulate MASP proteases of the lectin pathway of complement activation [26e28], the blood coagulation cascade [29] and the kallikrein/kinin system [30,31]. In patients with hereditary angioedema (HAE), a deficiency or dysfunction of C1Inh causes inappropriate activation of the kallikrein/kinin system. It has not been clarified whether SSL5 abrogates the suppression by C1Inh toward all these proteases, but SSL5 might have a major influence on a wide range of biological defense mechanisms. In the present experiments, the interaction between C1Inh and SSL5 was suggested to be mediated by sialic acid-containing glycans of C1Inh based on the following results: 1) the treatment of
We did not use samples collected from patients or animals in any of the experiments. Consent for publication All the authors have approved the manuscript and agree with its submission. Acknowledgments We are grateful to Dr. Misao Matsushita (Department of Applied Biochemistry, Tokai University) for his helpful discussion. We also thank Ms. Ayako Shimo, Ms. Keiko Otani, Ms. Yukari Yakawa, and Ms. Airi Watanabe (Hoshi University School of Pharmacy and Pharmaceutical Sciences) for their technical assistance. This work
Please cite this article as: T. Oku et al., Identification of human plasma C1 inhibitor as a target protein for staphylococcal superantigen-like protein 5 (SSL5), Biochemical and Biophysical Research Communications, https://doi.org/10.1016/j.bbrc.2018.12.026
6
T. Oku et al. / Biochemical and Biophysical Research Communications xxx (xxxx) xxx
was supported in part by Grants-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology of Japan and by Research Grants from the Shiono Wellness Foundation and the Japan Agency for Medical Research and Development (AMED). References [1] A.N. Spaan, B.G. Surewaard, R. Nijland, et al., Neutrophils versus Staphylococcus aureus: a biological tug of war, Annu. Rev. Microbiol. 67 (2013) 629e650. [2] T.J. Foster, Immune evasion by staphylococci, Nat. Rev. Microbiol. 3 (2005) 948e958. [3] V. Thammavongsa, H.K. Kim, D. Missiakas, et al., Staphylococcal manipulation of host immune responses, Nat. Rev. Microbiol. 13 (2015) 529e543. [4] J.D. Fraser, T. Proft, The bacterial superantigen and superantigen-like proteins, Immunol. Rev. 225 (2008) 226e243. [5] R.J. Williams, J.M. Ward, B. Henderson, et al., Identification of a novel gene cluster encoding staphylococcal exotoxin-like proteins: characterization of the prototypic gene and its protein product, SET1, Infect. Immun. 68 (2000) 4407e4415. [6] J. Bestebroer, M.J. Poppelier, L.H. Ulfman, et al., Staphylococcal superantigenlike 5 binds PSGL-1 and inhibits P-selectin-mediated neutrophil rolling, Blood 109 (2007) 2936e2943. [7] C.J. de Haas, C. Weeterings, M.M. Vughs, et al., Staphylococcal superantigenlike 5 activates platelets and supports platelet adhesion under flow conditions, which involves glycoprotein Ibalpha and alpha IIb beta 3, J. Thromb. Haemostasis 7 (2009) 1867e1874. [8] S. Itoh, E. Hamada, G. Kamoshida, et al., Staphylococcal superantigen-like protein 5 inhibits matrix metalloproteinase 9 from human neutrophils, Infect. Immun. 78 (2010) 3298e3305. [9] C. Kurisaka, T. Oku, S. Itoh, et al., Role of sialic acid-containing glycans of matrix metalloproteinase-9 (MMP-9) in the interaction between MMP-9 and staphylococcal superantigen-like protein 5, Microbiol. Immunol. 62 (2018) 168e175. [10] I. Jongerius, J. Kohl, M.K. Pandey, et al., Staphylococcal complement evasion by various convertase-blocking molecules, J. Exp. Med. 204 (2007) 2461e2471. [11] H. Chen, D. Ricklin, M. Hammel, et al., Allosteric inhibition of complement function by a staphylococcal immune evasion protein, Proc. Natl. Acad. Sci. U. S. A. 107 (2010) 17621e17626. [12] A.J. Laarman, M. Ruyken, C.L. Malone, et al., Staphylococcus aureus metalloprotease aureolysin cleaves complement C3 to mediate immune evasion, J. Immunol. 186 (2011) 6445e6453. [13] R. Langley, B. Wines, N. Willoughby, et al., The staphylococcal superantigenlike protein 7 binds IgA and complement C5 and inhibits IgA-Fc alpha RI binding and serum killing of bacteria, J. Immunol. 174 (2005) 2926e2933. [14] P.A. Ramsland, N. Willoughby, H.M. Trist, et al., Structural basis for evasion of IgA immunity by Staphylococcus aureus revealed in the complex of SSL7 with Fc of human IgA1, Proc. Natl. Acad. Sci. U. S. A. 104 (2007) 15051e15056. [15] N.S. Laursen, N. Gordon, S. Hermans, et al., Structural basis for inhibition of complement C5 by the SSL7 protein from Staphylococcus aureus, Proc. Natl. Acad. Sci. U. S. A. 107 (2010) 3681e3686.
[16] S. Itoh, E. Hamada, G. Kamoshida, et al., Staphylococcal superantigen-like protein 10 (SSL10) binds to human immunoglobulin G (IgG) and inhibits complement activation via the classical pathway, Mol. Immunol. 47 (2010) 932e938. [17] S. Itoh, R. Yokoyama, G. Kamoshida, et al., Staphylococcal superantigen-like protein 10 (SSL10) inhibits blood coagulation by binding to prothrombin and factor Xa via their gamma-carboxyglutamic acid (Gla) domain, J. Biol. Chem. 288 (2013) 21569e21580. [18] D. Patel, B.D. Wines, R.J. Langley, et al., Specificity of staphylococcal superantigen-like protein 10 toward the human IgG1 Fc domain, J. Immunol. 184 (2010) 6283e6292. [19] T. Oku, H. Soma, C. Kurisaka, et al., Generation of a monoclonal antibody against staphylococcal superantigen-like protein 5 (SSL5) that discriminates SSL5 from other SSL proteins, Monoclon. Antibodies Immunodiagn. Immunother. 37 (2018) 212e217. [20] T. Oku, S. Itoh, M. Okano, et al., Two regions responsible for the actin binding of p57, a mammalian coronin family actin-binding protein, Biol. Pharm. Bull. 26 (2003) 409e416. [21] T. Oku, S. Itoh, R. Ishii, et al., Homotypic dimerization of the actin-binding protein p57/coronin-1 mediated by a leucine zipper motif in the C-terminal region, Biochem. J. 387 (2005) 325e331. [22] X. Chen, J.L. Zaro, W.C. Shen, Fusion protein linkers: property, design and functionality, Adv. Drug Deliv. Rev. 65 (2013) 1357e1369. [23] T. Oku, Y. Ando, M. Ogura, et al., Development of splice variant-specific monoclonal antibodies against human alpha3 integrin, Monoclon. Antibodies Immunodiagn. Immunother. 35 (2016) 12e17. [24] M. Eckhardt, B. Gotza, R. Gerardy-Schahn, Mutants of the CMP-sialic acid transporter causing the Lec2 phenotype, J. Biol. Chem. 273 (1998) 20189e20195. [25] A.E. Davis III, P. Mejia, F. Lu, Biological activities of C1 inhibitor, Mol. Immunol. 45 (2008) 4057e4063. [26] M. Matsushita, S. Thiel, J.C. Jensenius, et al., Proteolytic activities of two types of mannose-binding lectin-associated serine protease, J. Immunol. 165 (2000) 2637e2642. [27] J.S. Presanis, K. Hajela, G. Ambrus, et al., Differential substrate and inhibitor profiles for human MASP-1 and MASP-2, Mol. Immunol. 40 (2004) 921e929. [28] L. Beinrohr, J. Dobo, P. Zavodszky, et al., MBL-MASPs and C1-inhibitor: novel approaches for targeting complement-mediated inflammation, Trends Mol. Med. 14 (2008) 511e521. [29] M. Cugno, I. Bos, Y. Lubbers, et al., In vitro interaction of C1-inhibitor with thrombin, Blood Coagul. Fibrinolysis 12 (2001) 253e260. [30] I. Gigli, J.W. Mason, R.W. Colman, et al., Interaction of plasma kallikrein with the C1 inhibitor, J. Immunol. 104 (1970) 574e581. [31] P.C. Harpel, M.F. Lewin, A.P. Kaplan, Distribution of plasma kallikrein between C-1 inactivator and alpha 2-macroglobulin in plasma utilizing a new assay for alpha 2-macroglobulin-kallikrein complexes, J. Biol. Chem. 260 (1985) 4257e4263. [32] S.C. Bock, K. Skriver, E. Nielsen, et al., Human C1 inhibitor: primary structure, cDNA cloning, and chromosomal localization, Biochemistry 25 (1986) 4292e4301. [33] K. Stavenhagen, H.M. Kayili, S. Holst, et al., N- and O-glycosylation analysis of human C1-inhibitor reveals extensive mucin-type O-glycosylation, Mol. Cell. Proteomics 17 (2018) 1225e1238.
Please cite this article as: T. Oku et al., Identification of human plasma C1 inhibitor as a target protein for staphylococcal superantigen-like protein 5 (SSL5), Biochemical and Biophysical Research Communications, https://doi.org/10.1016/j.bbrc.2018.12.026