Soluble N-Ethylmaleimide-Sensitive Factor Attachment Protein Receptor-Derived Peptides for Regulation of Mast Cell Degranulation

Vesicle-associated V-soluble N -ethylmaleimide-sensitive factor attachment protein receptor (SNARE) proteins and target membrane-associated T-SNAREs (syntaxin 4 and SNAP-23) assemble into a core trans -SNARE complex that mediates membrane fusion during mast cell degranulation. This complex plays pivotal roles at various stages of exocytosis from the initial priming step to fusion pore opening and expansion, finally resulting in the release of the vesicle contents. In this study, peptides with the sequences of various SNARE motifs were investigated for their potential inhibitory effects against SNARE complex formation and mast cell degranulation. The peptides with the sequences of the N-terminal regions of vesicle-associated membrane protein 2 (VAMP2) and VAMP8 were found to reduce mast cell degranulation by inhibiting SNARE complex formation. The fusion of protein transduction domains to the N-terminal of each peptide enabled the internalization of the fusion peptides into the cells equally as efficiently as cell permeabilization by streptolysin-O without any loss of their inhibitory activities. Distinct subsets of mast cell granules could be selectively regulated by the N-terminal-mimicking peptides derived from VAMP2 and VAMP8, and they effectively decreased the symptoms of atopic dermatitis in mouse models. These results suggest that the cell membrane fusion machinery may represent a therapeutic target for atopic dermatitis.

Cloning, Expression, and Purification of Recombinant SNARE Proteins Total RNA was obtained from RBL-2H3 cells using TRIzol ® Reagent (#15596026, Invitrogen, Carlsbad, CA, USA) and served as a template for cDNA synthesis with M-MLV reverse transcriptase (M1705, Promega, Madison, WI, USA) according to the manufacturer’s protocol. The cDNAs amplified for rat Syn4, SNAP-23, VAMP2, VAMP4, VAMP7, and VAMP8 using each primer pair ( 4 ), were cloned into the pGEX-4T 1 vector. The mast cell SNARE proteins [Syn4 (amino acids 1–298), SNAP-23 (amino acids 1–210), full-length VAMP2 (VpF, amino acids 1–116), VAMP4 (amino acids 1–141), and VAMP8 (amino acids 1–100)] were expressed in Escherichia coli Rosetta(DE3)pLysS (#70956, Novagen, Darmstadt, Germany) and purified as glutathione- S -transferase-tagged fusion proteins as reported previously ( 8 ). Briefly, the E. coli cells were grown at 37°C in Luria–Bertani medium (#244610, BD Biosciences, San Diego, CA, USA) and the expression of the recombinant protein was induced by treating the cells with 0.3 mM isopropyl β- d -thiogalactoside when their optical density at 600 nm reached 0.8. Then, the cells were harvested by centrifugation at 8,000 g for 10 min and lysed by sonication (45% amplitude, 1.5 min net sonication, 1 s on–1 s off). The lysate was clarified by centrifugation (10,000 g , 20 min, 4°C), and bound to glutathione agarose beads (G4510, Sigma-Aldrich, St. Louis, MO, USA). Following a thorough washing step, the glutathione- S -transferase-tagged SNAREs were cleaved using thrombin (27-0846-01, GE Healthcare, Pittsburgh, PA, USA). The purity of all proteins, as assessed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis using 12.5% polyacrylamide gels, was greater than 90%.

To prepare the unilamellar liposomes, a 50-mM lipid mixture of palmitoyl-2-oleoylphosphatidylcholine (POPC) (#850457, Avanti Polar Lipids, Alabaster, AL, USA):1,2-dioleoyl-phosphatidylserine (DOPS) (#840035, Avanti Polar Lipids) at 65:35 mol% was dried in a glass tube under a gentle stream of nitrogen gas and then exposed to a vacuum overnight. The resultant lipid film was resuspended in dialysis buffer (25 mM HEPES and 100 mM KCl, pH 7.4) by vortexing, followed by an extrusion step with polycarbonate membrane filters having a pore size of 100 nm (#610005, Avanti Polar Lipids). To prepare proteoliposomes, binary T-SNAREs, which were pre-formed by mixing Syn4 and SNAP-23 at room temperature for 60 min, were incubated with 50 mM unilamellar liposomes for 15 min at the same temperature, resulting in a 50:1 lipid/protein molar ratio ( 9 ). For the preparation of V-SNARE proteoliposomes, a 10-mM premixed lipid solution [POPC:DOPS: N -(7-nitro-2-1,3-benzoxadiazol-4-yl (NBD))-phosphatidylserine (#810198, Avanti Polar Lipids):rhodamine-phycoerythrin (#810150, Avanti Polar Lipids) = 62:35:1.5:1.5 (molar ratio)] was used for the preparation of fluorescent liposomes in the same manner as above. The prepared liposomes were then mixed with VAMP2 for 15 min at room temperature. The liposome/protein mixture was diluted twice to bring the concentration of n -octyl glucoside (O8001, Sigma-Aldrich) below the critical micelle concentration. After overnight dialysis against 1L dialysis buffer at 4°C to remove the detergent, SM-2 bio-beads (#1523920, Bio-Rad Laboratories, Hercules, CA, USA) were added to the sample to remove any remaining n -octyl glucoside. The solution was then centrifuged at 10,000 × g to remove the protein/lipid aggregates, and the final lipid concentrations of both the T- and V-liposomes were adjusted to 1 mM with a lipid-to-protein ratio of 50:1. A total lipid mixing assay using the T- and V-liposomes was performed as described previously with minor modifications ( 8 ). Briefly, peptide inhibitors were applied to the T-vesicles at the indicated concentrations and then the T-vesicles were mixed with the V-vesicles in a 384-well plate for the initiation of fusion, with a T- to V-vesicle volume ratio of 9:1. The fusion was monitored using the dequenching of NBD fluorescence signal (excitation 465 nm/emission 530 nm). After at least 80 min of fusion, the maximum NBD signal was obtained by adding 0.1% Triton X-100™. All the lipid-mixing experiments were performed at 37°C.

Seven-week-old NC/Nga mice were purchased from the Shizuoka Laboratory Animal Center (Shizuoka, Japan) and allowed to adapt to the environment for a week prior to use. The procedures for the induction of atopic dermatitis-like symptoms in mice using 1-chloro-2,4-dinitro-benzene (DNCB) have been well described previously ( 27 ). Briefly, the dorsal hair of the mice was shaved 2 days before DNCB administration. For the induction of atopic dermatitis, 1% DNCB was applied to the dorsal skin of the mice twice weekly for 7 days and then 0.2% DNCB was applied thrice weekly for 21 days. The mice were divided into the following seven treatment groups ( n = 10 per group): 0.3 and 3% Vp2N/day, 0.3 and 3% Vp8N/day, 0.1% dexamethasone/day, and vehicle only, dorsally administered from 9 to 13 weeks of age. As a negative control for atopic dermatitis, NC/Nga mice that were not treated with DNCB were maintained under pathogen-free conditions. Laboratory animal breeding management was based on the “Guide for the Care and Use of Laboratory Animals,” and all experiments were approved by the Institutional Animal Care and Use Committees of Gyeonggi Institute of Science & Technology.

Peptides Having the Sequences of Various SNARE Motifs To search for potent peptide inhibitors that can interfere with SNARE complex formation and thereby reduce mast cell degranulation, seventeen 17-mer peptides were synthesized of which sequences were patterned after the N-terminal (N), middle (M), and C-terminal (C) regions of the SNARE motifs of SNAP-23 (SN23 and SC23), Syn4, and VAMPs (Vp2, 4, 7, and 8) (Table 1 ; Figure 1 ). Those four SNARE proteins were selected as the template sequences because the activity of their ternary complexes was validated in previous studies using the lipid mixing assay ( 4 , 8 , 21 ). Additionally, the full-length soluble SNARE motifs of VAMP2, VAMP4, VAMP7, and VAMP8, which are known to be present in mast cell granules, were expressed and purified from E. coli . The PTDs were fused to the N-terminal regions of the synthetic peptides to provide the peptides with the ability to cross cell membranes ( 22 ). PTDs are small peptides that are used to carry proteins, peptides, nucleic acids, nanoparticles, and other types of cargo across the plasma membrane into cells ( 28 – 30 ). Shorter peptides with six or eight amino acid residues were not tested because they did not inhibit membrane fusion by neuronal SNARE proteins even at 200 µM concentration ( 9 ). Table 1 Amino acid sequences of synthetic peptides derived from mast cell soluble N -ethylmaleimide-sensitive factor attachment protein receptor (SNARE) motifs. Name Sequence SN23N RAHQVTDESLESTRRIL SN23M KTITMLDEQGEQLNRIE SN23C DQINKDMREAEKTLTEL SC23N EDEMEENLTQVGSILGN SC23M DMGNEIDAQNQQIQKIT SC23C DTNKNRIDIANTRAKKL Syn4N SARHSEIQQLERTIREL Syn4M FLATEVEMQGEMINRIE Syn4C DYVERGQEHVKIALENQ Vp2N RRLQQTQAQVDEVVDIM Vp2M VNVDKVLERDQKLSELD Vp2C QFETSAAKLKRKYWWKN Vp8N LQSEVEGVKNIMTQNVE Vp8M RGENLDHLRNKTEDLEA Vp8C EHFKTTSQKVARKFWWK Vp4N KIKHVQNQVDEVIDVMQ Vp7N QSLDRVTETQAQVDELK Protein transduction domain (PTD) YARVRRRGPRRGGG Seventeen 17-mer peptides representing the N-terminal, middle, and C-terminal regions of the SNARE motifs of SN23, syntaxin 4, Vp2, Vp8, Vp4, and Vp7 were synthesized. PTDs were added to the N-terminus of each peptide to facilitate their penetration into cells . Figure 1 Design of mast cell soluble N -ethylmaleimide-sensitive factor attachment protein receptor (SNARE) motif-patterned peptides. Peptides patterned after the α-helical regions of the SNARE motifs of SN23 (green), syntaxin 4 (red), and VAMPs (blue) were designed. The names of the peptides are shown in the motif diagram, and the residue number in the motif corresponding to each peptide is indicated above each diagram. The amino acid sequences of the synthetic peptides are shown in Table 1 . The soluble domains of various VAMP proteins are also indicated. TMD, transmembrane domain.

A lipid-mixing assay was performed using SNARE-reconstituted liposomes ( 37 , 38 ). Proteoliposomes containing either V-SNARE (Vp2 or Vp8) and both T-SNAREs (SN23 and Syn4) were prepared to compare the abilities of those SNARE protein combinations to merge membranes. Both the tested combinations of SNARE proteins (SN23-Syn4-Vp2 and SN23-Syn4-Vp8) efficiently mediated membrane fusion, as indicated by the increase in the fluorescence intensity (Figure 3 A), confirming that both of these SNARE complexes are active. When Vp2S or Vp8S was added to the reaction mixture, the membrane fusion mediated by SN23-Syn4-Vp2 or SN23-Syn4-Vp8 complexes was almost completely inhibited by both proteins (Figure 3 B). The similarity between the activities of the two SNARE complexes containing each full-length soluble SNARE is consistent with the result that both β-hexosaminidase and histamine release were inhibited by both of those SNARE proteins (Figure 2 D). Figure 3 Inhibitory effects of the synthetic peptides on SNARE-driven membrane fusion. (A) Lipid mixing assay of T-liposomes containing syntaxin 4 (Syn4) and SN23 together with V-liposomes prepared with Vp2 (black) or Vp8 (gray). The assay was monitored by measuring the dequenching of NBD fluorescence and the data were plotted as a function of time. The maximum NBD signal was obtained by adding 0.1% Triton X-100™. (B) The soluble N-ethylmaleimide-sensitive factor attachment protein receptor (SNARE)-mediated membrane fusion was completely blocked by the cytoplasmic domains of Vp2 (Vp2S, red) and Vp8 (Vp8S, green) at the concentration of 10 µM. (C) The inhibitory effects of synthetic peptides on lipid mixing using mast cell SNAREs including Vp2 (black) and Vp8 (gray). Each bar represents the mean ± SD of three independent experiments. The statistical significance of differences was evaluated by analysis of variance (* p < 0.05). (D) The curves represent the lipid mixing driven by SN23-Syn4-Vp2 or SN23-Syn4-Vp8 in the presence of the most effective peptides (Syn4N, Vp2N, and Vp8N). In contrast to the results described above, several peptides showed significantly different inhibitory activities against the SNARE complexes of SN23-Syn4-Vp2 and SN23-Syn4-Vp8 (Figure 3 C). While most peptides inhibited membrane fusion by only 0–20%, the peptides SC23N, Syn4N, Vp2N, and Vp8N potently inhibited membrane fusion with specificities for different SNAREs. The peptides derived from the N-terminal region of the T-SNARE proteins SC23N and Syn4N inhibited membrane fusion by both the SN23-Syn4-Vp2 and SN23-Syn4-Vp8 complexes (Figures 3 C,D). Vp2N efficiently inhibited the membrane fusion mediated by SN23-Syn4-Vp2 but not that mediated by SN23-Syn4-Vp8. In contrast, the inhibition of membrane fusion by Vp8N and Vp8M was specific to the SN23-Syn4-Vp8 complex. These results clearly indicated that the Vp2N and Vp8N peptides derived from the N-terminal region of the V-SNAREs VAMP2 and VAMP8 specifically inhibited the membrane fusion processes driven by VAMP2 and VAMP8, respectively. Because the N-terminal nucleation step of SNARE complex formation is the rate-limiting step of SNARE-driven membrane fusion ( 39 ), the membrane fusion driven by the neuronal SNARE complex (SNAP-25-VAMP2-syntaxin 1a) was most efficiently inhibited when N-terminal binding flavonoids (cyanidin and delphinidin) ( 8 , 40 ) or N-terminal-mimicking peptides were present ( 9 ). It is also possible that the efficiency and specificity of Vp2N and Vp8N against only their cognate SNARE complexes could be attributed to the rate-limiting N-terminal nucleation of the SNARE complexes comprising SN23-Syn4-Vp8 and SN23-Syn4-Vp2. The peptides Syn4N, SC23N, Vp2N, Vp8N, SN23C, and Vp8C were added to the reaction mixture containing the T-vesicle and each V-vesicle before or after preincubation at 4°C, which represents the condition in which the N-terminal SNARE proteins are partially zippered without any lipid mixing ( 38 ). When the peptides were added before preincubation, the inhibitory effects of the peptides were consistent with the above results, except that the inhibitory effects of the N-terminal-mimicking peptides increased [Figure 4 A (gray bars) and Figure 3 C]. The addition of SC23N and Syn4N derived from the T-SNARE proteins inhibited the membrane fusion induced by both SN23-Syn4-Vp2 and SN23-Syn4-Vp8 when the peptides were applied before preincubation. In contrast, Vp2N and Vp8N inhibited the membrane fusion driven only by their cognate SNARE complexes before preincubation. SN23C and Vp8C did not inhibit membrane fusion regardless of preincubation (Figure 4 A). However, when the most effective peptides (Vp2N, Vp8N, Syn4N, and SC23N) were added after preincubation, their inhibitory activities significantly diminished (Figure 4 A). Thus, the N-terminal-mimicking peptides inhibit membrane fusion by hindering the N-terminal nucleation step of SNARE complex formation. Figure 4 The inhibitory effect of synthetic peptides on th

Botulinum neurotoxin is now widely applied in the cosmetic field and for the treatment of various neurological diseases such as strabismus, blepharospasm, hemifacial spasm, cervical dystonia, and many other indications ( 46 ). To realize new BoNT-like compounds for use in the cosmetic industry, there have been numerous attempts to mimic the efficacy of BoNT using peptides ( 9 , 47 ) and plant-derived small molecules ( 4 , 8 , 10 , 21 , 48 ) by regulating neuronal SNARE complex formation. Acetyl hexapeptide-3 (Argireline ® ) is a commercialized synthetic peptide that is patterned from the N-terminal end of the SNAP-25 protein and inhibits SNARE complex formation as well as the release of catecholamine, adrenaline, and noradrenaline ( 49 ). The elongated form of Argireline ® , acetyl octapeptide-3 (SNAP-8), has also been developed for reducing wrinkles in aging skin ( 50 ). In addition, the N-terminal-truncated SNAP-25 that is cleaved by BoNT/A or BoNT/E has been found to inhibit neurotransmitter release ( 16 ). Several peptides identified by library deconvolution that do not mimic the SNARE domain have also been shown to have inhibitory effects on the assembly of the SNARE complex ( 51 – 53 ). Since the cell membrane fusion machinery has been clearly shown to be a versatile therapeutic target, an engineered BoNT that cleaves both SNAP-23 and SNAP-25 was previously developed ( 54 ). Although the therapeutic potential of that engineered BoNT has not been demonstrated yet, it is expected to have the potential for the treatment of human hypersecretion disorders such as atopic dermatitis, which is associated with a massive degranulation of mast cells. Thus, it was also plausible that a similar strategy used for the neuronal SNAREs can be applicable to mast cell SNAREs using peptides. Given that the ternary SNARE complexes of SN23-Syn4-Vp2 and SN23-Syn4-Vp8 are critically involved in the mast cell degranulation process ( 3 , 19 , 41 ), the inhibition of SNARE complex formation may provide an opportunity for therapeutic interventions against atopic dermatitis. Our challenge in the present study was to control the SNARE-mediated degranulation of mast cells and thereby regulate atopic dermatitis using synthetic peptides. The inhibitory effects of various peptides patterned after SNARE motifs against mast cell degranulation were investigated. In vitro experiments demonstrated a marked inhibition of the IgE-stimulated release of β-hexosaminidase and histamine by N-terminal-mimicking peptides through interference with Vp2- or Vp8-associated membrane fusion. This result is consistent with the previous results, where N-terminal-mimicking peptides of neuronal SNARE proteins were the most potent inhibitors of neurotransmitter release ( 9 ). It is because N-terminal nucleation is the rate-limiting step of SNARE complex formation. The peptides mimicking other regions of SNARE proteins showed very low inhibitory activity against SNARE-driven fusion and mast cell degranulation, consistently to those for neuronal SNARE complex. This low activity is attributed to the folding pathway of SNARE complex. C-terminal fragment of Vp2 (aa 49-96) bound to the binary acceptor complex can be readily removed by native Vp2 even without loss of speed (N- to C-Terminal SNARE Complex Assembly Promotes Rapid Membrane Fusion). Thus, 17-mer M- or C-peptides that bind to pre-formed binary complex are likely to be peeled off by native Vp2 and cannot inhibit full SNARE zippering. The only exception reported until now is myricetin which stops SNARE zippering in the middle ( 21 ). Because SNARE complexes have been shown to zipper in three distinct stages ( 52 ), the folding intermediate of ternary complex seems to expose a binding pocket near the ionic zeroth layer ( 21 ). The efficacy of N-terminal-mimicking peptides were confirmed in NC/Nga mice model with atopic dermatitis-like skin lesions. Although the detailed molecular mechanisms underlying the therapeutic and side effects of the peptides remain to be elucidated, our findings suggest that peptide inhibitors alleviate atopic dermatitis-like skin lesions in NC/Nga mice by targeting SNARE-mediated membrane fusion. Therefore, we propose that N-terminal-mimicking peptides may represent a promising topical therapeutic candidate for the control of atopic dermatitis. Mast cells are multifunctional cells and play an important role in allergic and inflammatory responses. They are activated by a series of stimuli, releasing and producing inflammatory mediators ( 55 , 56 ). The rat basophilic leukemia cells, RBL-2H3, are a tumor analog of mast cells, which can be activated by the IgE–antigen complex. In this study, RBL-2H3 cells were used to investigate the anti-allergic activity of SNARE-mimicking peptides. We note the limitations of RBL-2H3 cells for a model of mast cell mediator release because RBL-2H3 cells have been shown to have similarities to basophils in some aspects rather than other histamine-rele

YY and D-HK contributed conception and design of the study; YY, BK, J-MO, JH, and D-HK performed experiments; YJ, J-BP, J-MO, and JH performed the statistical analysis; YY and D-HK wrote the first draft of the manuscript; YY, BK, JC, and D-HK wrote sections of the manuscript. All authors contributed to manuscript revision, read and approved the submitted version.

The Supplementary Material for this article can be found online at https://www.frontiersin.org/articles/10.3389/fimmu.2018.00725/full#supplementary-material . Click here for additional data file.