- Email: [email protected]
Generation of a novel artificial TrkB agonist, BM17d99, using T7 phage-displayed random peptide libraries
Biochemical and Biophysical Research Communications xxx (2017) 1e6
Contents lists available at ScienceDirect
Biochemical and Biophysical Research Communications journal homepage: www.elsevier.com/locate/ybbrc
Generation of a novel artificial TrkB agonist, BM17d99, using T7 phage-displayed random peptide libraries Toshiyuki Ohnishi 1, Kotaro Sakamoto*, 1, Asano Asami-Odaka, Kimie Nakamura, Ayako Shimizu, Takashi Ito, Taiji Asami, Tetsuya Ohtaki, Hiroshi Inooka Pharmaceutical Research Division, Takeda Pharmaceutical Company Limited, 26-1, Muraoka-Higashi 2-chome, Fujisawa, Kanagawa 251-8555, Japan
a r t i c l e i n f o
a b s t r a c t
Article history: Received 21 December 2016 Accepted 29 December 2016 Available online xxx
Tropomyosin receptor kinase B (TrkB) is a known receptor of brain-derived neurotrophic factor (BDNF). Because it plays a critical role in the regulation of neuronal development, maturation, survival, etc., TrkB is a good target for drugs against central nervous system diseases. In this study, we aimed to generate peptidic TrkB agonists by applying random peptide phage display technology. After the phage panning against recombinant Fc-fused TrkB (TrkB-Fc), agonistic phages were directly screened against TrkBexpressing HEK293 cells. Through subsequent screening of the first-hit BM17 peptide-derived focus library, we successfully obtained the BM17d99 peptide, which had no sequence similarity with BDNF but had TrkB-binding capacity. We then synthesized a dimeric BM17d99 analog peptide that could phosphorylate or activate TrkB by facilitating receptor homodimerization. Treatment of TrkB-expressing HEK293 cells with the dimeric BM17d99 analog peptide significantly induced the phosphorylation of TrkB, suggesting that homodimerization of TrkB was enhanced by the dimeric peptide. This report demonstrates that our approach is useful for the generation of artificial peptidic agonists of cell surface receptors. © 2016 Elsevier Inc. All rights reserved.
Keywords: Peptide Agonist TrkB BDNF Phage display
1. Introduction Tropomyosin receptor kinase B (TrkB) and its natural ligand, brain-derived neurotrophic factor (BDNF), which are expressed in many regions of the brain, are involved in psychiatric disorders (i.e., depression, schizophrenia) and neurodegenerative diseases (i.e., Alzheimer's disease, Huntington's disease) [1e3]. Despite these diverse pharmacological effects, BDNF has only a short half-life in the blood and poor central transition [4], making it difficult to use intrinsic BDNF as a therapeutic agent. Therefore, alternative molecules that mimic BDNF function would serve as novel innovative central nervous system (CNS) drugs. In recent years, small molecules, such as the LM22A series, have been reported as TrkB agonists [5]. However, these compounds did not show TrkB agonistic activity in our experimental system. The TrkB agonistic antibodies reported by Wyeth Research and Pfizer
Abbreviations: TrkB, tropomyosin-receptor-kinase B; BDNF, brain-derived neurotrophic factor; RTK, receptor tyrosine kinase. * Corresponding author. E-mail address: [email protected] (K. Sakamoto). 1 The first two authors contributed equally to this work.
Inc., respectively [6e8], showed remarkable specificity and agonistic activity, but they had no central transition ability because of their higher molecular weight. In contrast to these molecules, peptidic agonists of TrkB should possess binding specificity and the potential for CNS delivery. Since TrkB is a member of the receptor tyrosine kinases (RTKs), receptor dimerization is the key modification for its activation. The most successful example of an artificial peptidic agonist to RTK is the erythropoietin (EPO)-mimicking peptide Hematide [9e11], which was discovered through random peptide phage display technology and has no sequence similarity to EPO. Phage display is an extremely efficient method for the isolation of novel peptides that bind to target molecules [12,13]. Ever since the artificial peptidic agonist to RTKs was generated, several key achievements have been made: (1) discovery of a monomeric peptide with high affinity to the receptor, and (2) development of a potent dimerized peptide using a chemical linker, such as Lys (diamino acid), Asp (dicarboxylic acid), or bismaleimide. Although the process appears simple, it is actually quite difficult. Since the generation of Hematide, only a few artificial peptidic agonists to RTKs have been reported [14e16]. In this study, we aimed to obtain novel peptides that are
http://dx.doi.org/10.1016/j.bbrc.2016.12.186 0006-291X/© 2016 Elsevier Inc. All rights reserved.
Please cite this article in press as: T. Ohnishi, et al., Generation of a novel artificial TrkB agonist, BM17d99, using T7 phage-displayed random peptide libraries, Biochemical and Biophysical Research Communications (2017), http://dx.doi.org/10.1016/j.bbrc.2016.12.186
2
T. Ohnishi et al. / Biochemical and Biophysical Research Communications xxx (2017) 1e6
agonistic to TrkB. First, TrkB-Fc was constructed as a pseudoreceptor dimer for the affinity screening of T7 phage-displayed random peptide libraries. Next, the potential agonistic phages were directly screened against TrkB-expressing HEK293 cells. As a result, an agonistic phage (BM17) was successfully isolated, whose synthetic monomeric peptide exhibited significant agonistic activity in cell-based assays. Through affinity enhancement and chemical modification of the BM17 sequence, the potent TrkB agonist BM17d99 dimeric peptide was finally generated. 2. Materials and methods 2.1. Materials Protein constructs of the extracellular domain of human TrkB (Met1-His430) fused to the Fc portion of immunoglobulin G (TrkBFc) and human BDNF were produced internally, using a proprietary vector and expression in FreeStyle 293 cells. Human TrkA-Fc (1056TK), human TrkC-Fc (375-TC), and human p75NTR-Fc (367-NR) were purchased from R&D Systems (Minneapolis, MN, USA). A series of phage-derived peptides were chemically synthesized by employing Fmoc protection chemistry and were purified on a reversed phase column. Primary mouse cortical neurons were isolated from the cerebral cortex of E15 mice through use of a dissociation solution (MB-X0802; Sumitomo Bakelite Co., Ltd, Tokyo, Japan) and cultured in Neurobasal medium (21103049; Thermo Fisher Scientific, Waltham, MA, USA) containing 2% B27 supplement, GlutaMAX, and antibiotics. A human cell line stably expressing TrkB was constructed by transducing the human TrkB (variant C; GenBank No. NM_001018064)-expressing pcDNA3.1 vector to HEK293 cells and selecting for antibiotic resistance using geneticin (G418). These cells were maintained in Dulbecco's modified Eagle's medium (DMEM) containing 10% fetal bovine serum (FBS) and G418. 2.2. Phage library construction and panning The T7 phage-displayed random peptide libraries, which were generated by using mixed oligonucleotides as template DNA, were constructed by using a T7Select 10-3 vector from Merck (Darmstadt, Germany), according to methods described previously [12,13]. The recombinant Fc-fused protein was immobilized to Dynabeads Protein A/G (Invitrogen, Carlsbad, CA, USA) with 1.0% casein in phosphate-buffered saline (PBS). After washing with PBS containing 0.1% Tween 20 (PBST), the beads were incubated with the phage libraries for 1 h and then washed with PBST. The bound phages were eluted with 1% sodium dodecyl sulfate and then used to infect log-phase Escherichia coli BLT5615 cells (Merck) for phage amplification. After bacteriolysis, the phages were recovered from the culture supernatant by centrifugation and polyethylene glycol precipitation. The recovered phages were dissolved in PBS for use in the next round of panning. 2.3. Evaluation of phage- or synthetic peptide-binding to recombinant proteins by ELISA The wells of a Nunc Maxisorp microplate (460e518) were coated with anti-human Fc goat polyclonal antibody (109-005-098; Jackson ImmunoResearch, West Grove, PA, USA), followed by blocking with 1.0% casein in PBS. Fc-fused proteins were captured by the antibody, and the phage solution or biotinylated peptide solution was then added to the wells. After washing with PBST, the bound phages or peptides were detected by using horseradish peroxidase (HRP)-conjugated anti-T7 antibody (Merck) or HRP-conjugated streptavidin (Vector Laboratories, Burlingame, CA, USA),
respectively. The amounts of HRP in the wells were measured by the detection reagent tetramethylbenzidine (Wako Pure Chemical Industries, Osaka, Japan) or by a chemiluminescent reagent (Wako). 2.4. Chemical synthesis of dimeric peptide The peptide and triethylamine (12 equivalents) were dissolved in dimethylformamide (DMF), following which 0.1 M 4dimethylaminopyridine in DMF (0.1 equivalents) and the chemical linker Fmoc-bAsp(OSu)2 (bis(2,5-dioxopyrrolidin-1-yl) 3((((9H-fluoren-9-yl) methoxy) carbonyl) amino) pentanedioate) (M02722; Watanabe Chemical Instituted, Hiroshima, Japan) in DMF (0.5 equivalents) were added. The reaction mixture was mixed for 2 h, following which diethylamine was added to remove the Fmoc. After mixing for 15 min, 0.1 M acetic acid was added to terminate the reaction. The mixture was filtered through polytetrafluorethylene membrane filters, and purified by reversed phase highperformance liquid chromatography (Phenomenex C18 column, 4.6 4 150 mm). Pure fractions were collected and lyophilized to yield the dimeric peptide as a white powder. 2.5. Evaluation of agonistic activity of phages or synthetic peptides in cell-based assay Human TrkB-expressing HEK293 cells were plated on poly-DLys-coated plates in DMEM containing 10% FBS and G418. After exchanging the medium to 0.1% BSA-containing DMEM, phages or synthetic peptides diluted with 0.05% CHAPS-containing DMEM were added to the wells and allowed to incubate for 30 min. Then, the phage solution or peptide solution was removed, and the cells were lysed with RIPA buffer (pH 7.5; 50 mM Tris-HCl, 1%Triton X100, 150 mM NaCl, and 5 mM ethylenediaminetetraacetic acid) containing a protease inhibitor cocktail and phosphatase inhibitors (Roche, Basel, Switzerland). The cell lysates were electrophoresed and then transferred to a polyvinylidene fluoride membrane by the TransBlot Turbo system (Bio-Rad, Hercules, CA, USA.). For detection of phosphorylated TrkB or extracellular signal-regulated kinases 1 and 2 (Erk1/2), anti-pY516 rabbit monoclonal antibody (Cell Signaling Technology, Beverly, MA, USA) or anti-p44/42 MAPK (T202/Y204) rabbit polyclonal antibody (Cell Signaling Technology) were used, respectively. For the secondary antibody, IRDye 800CWlabeled anti-rabbit IgG antibody (1:800; #611-131-122; Rockland Immunochemicals Inc., Limerick, PA, USA) was used, and the band intensity was measured at the extension and emission wavelengths of 780 and 820 nm, respectively, with the use of the Odyssey Imaging System (Li-COR, Lincoln, NE, USA). When performing the experiments with primary mouse cortical neurons, Neurobasalbased culture medium was used. 3. Results 3.1. Recombinant TrkB-Fc binding phages were isolated from T7 phage libraries To obtain TrkB-binding peptides, random peptide-displaying T7 phage libraries were panned against TrkB-Fc. At the 5th round of panning, TrkB-binding polyclonal phages were concentrated (Fig. 1A), and a total of 475 clones were picked to test their TrkBbinding activity. In combination with sequence analysis, 11 independent sequences having TrkB-binding affinity were obtained (Table 1). These clones shared consensus motifs with one another and had no sequence similarity with BDNF. Although these 11 phage clones bound selectively to TrkB-Fc (the results from representative phage clone BM17 are shown in Fig. 1B), they had no competitiveness with BDNF in TrkB binding (data not shown).
Please cite this article in press as: T. Ohnishi, et al., Generation of a novel artificial TrkB agonist, BM17d99, using T7 phage-displayed random peptide libraries, Biochemical and Biophysical Research Communications (2017), http://dx.doi.org/10.1016/j.bbrc.2016.12.186
T. Ohnishi et al. / Biochemical and Biophysical Research Communications xxx (2017) 1e6
3
Fig. 1. Biopanning of random peptide-displaying phage libraries against TrkB-Fc. (A) Binding activity of polyclonal phages to TrkB-Fc in ELISA. The Fc-fusion proteins were immobilized by anti-Fc antibody, and the phage binding was detected with horseradish peroxidase-conjugated anti-T7 phage antibody. (B) Binding specificity of monoclonal phage BM17. The phage binding activity to TrkB-Fc, TrkA-Fc, TrkC-Fc, and p75NTR-Fc was evaluated. The number of phages was expressed as plaque-forming units (pfu) per well.
Table 1 TrkB-binding sequences obtained from T7 phage display. Cluster
Name
Sequence
1 1 1
BM17 BM40 BM36
NVRPRICRVRKWTLCF PAKHLCLLAANRVRRSWQCL PRHKKCFRVLRACIEY
2 2 2
BM97 BM124 BM101
SDTCWKFYRGWPRRCRSN HAWCFPGPSLWPRRCRNN KSWCRQGFWPIRCSTT
3 3
BM15 BM64
GPWKLSFL GPWNYRSYKFWSKTAK
Singleton Singleton Singleton
BM07 BM74 BM100
AWPYQNWSFWTSRFSD NIPSIRDCIPIGCLQYAFWR KRACKHSHGIWCTKW
panning against TrkB-Fc, sequence analysis revealed that the phages with no TrkB-binding activity had a Phe16 deletion or Cterminal elongation (data not shown). Based on the above results, another BM17-derived T7 phage library was reconstructed, which displayed peptides having several mutations within the BM17 sequence without changing the chain length. The screening showed several phage clones with increased TrkB-binding activity (Table 2). In particular, phage BM17d99 (KSLPRMCRVRKWRLCF) had the highest affinity, being 10.8-times higher than that of the parental BM17 phage (Table 2). The EC50 binding values for the BM17 and BM17d99 synthetic peptides were 17 nM and 0.45 nM, respectively (Fig. 3), demonstrating BM17d99 to have a 38-times higher binding affinity than the parental BM17 peptide. The relative affinity between the BM17 and BM17d99 sequences was consistent in both peptide and phage forms.
Underlines indicate conserved or similar amino acid positions.
3.4. Dimeric BM17d99 peptide analog activated cell surface TrkB 3.2. BM17 sequence showed agonistic activity both in phage form and as a synthetic peptide Next, we directly evaluated the agonistic activity of the 11 phage clones in a cell-based assay. Three (BM17, BM100, and BM124) out of the 11 clones induced phosphorylation of both TrkB and its downstream Erk1/2 (Fig. 2A), even though the concentration of the particular peptide displayed in 2.5 1011 pfu/mL of phage suspension was very low (approximately 420 pM, assuming one phage particle corresponds to one peptide molecule). By additionally selecting two clones (BM36 and BM40) that have sequence similarity to BM17 to examine the structure-activity relationships, a total of 5 sequences (BM17, BM36, BM40, BM100, and BM124) were chemically synthesized. Among these, only peptide BM17 (NVRPRICRVRKWTLCF) exhibited significant phosphorylation of TrkB in mouse primary neurons (Fig. 2B). 3.3. BM17d99 sequence was 10-times more potent than the parental BM17 sequence For affinity enhancement, a BM17-derived T7 phage library (NVRPRICRVRKWTLCXXXXX) was constructed, where 5 random amino acid residues were added at the C-terminus of BM17. After
As described above, the monomeric BM17 peptide induced phosphorylation of TrkB, even though its agonistic activity was very weak (EC50 ¼ 104 M). Since receptor homodimerization is needed for TrkB activation, we hypothesized that dimerization of the BM17d99 peptide, by conjugation of its N-terminus to the dicarboxylic acid-linker bAsp, might increase its agonistic activity. Prior to examining this hypothesis, Lys1 and Lys11, both with reactive amino groups in their side chains, were substituted to Asn and Arg, respectively, to avoid undesired conjugate reactions. The resultant monomeric BM17d99(K1N/K11R) peptide showed an approximately 5-fold reduction in TrkB-Fc-binding activity in comparison with BM17d99, as demonstrated by ELISA (data not shown). However, the conjugated dimeric BM17d99(K1N/K11R) peptide demonstrated an almost 10,000-fold increase in its agonistic activity (EC50 ¼ 10 nM) compared with that of the monomeric BM17 peptide, suggesting that peptide dimerization by linkage had successfully led to the cooperative binding of the BM17d99(K1N/K11R) moieties to the TrkB monomers, with subsequent TrkB homodimerization (Fig. 4). Moreover, intracerebroventricular administration of the dimeric BM17d99(K1N/K11R) peptide to C57BL/6NCrl mice significantly induced TrkB phosphorylation in the hypothalamus, suggesting that our peptide could activate cell surface TrkB not only in vitro but also in vivo (Supplementary Fig. S1).
Please cite this article in press as: T. Ohnishi, et al., Generation of a novel artificial TrkB agonist, BM17d99, using T7 phage-displayed random peptide libraries, Biochemical and Biophysical Research Communications (2017), http://dx.doi.org/10.1016/j.bbrc.2016.12.186
4
T. Ohnishi et al. / Biochemical and Biophysical Research Communications xxx (2017) 1e6
Fig. 2. Evaluation of TrkB-agonistic activity of phages and synthetic peptides. (A) Phosphorylation of TrkB and Erk1/2 by phage particles (2.5 1011 pfu/mL) using TrkBexpressing HEK293 cells. (B) Phosphorylation of TrkB by synthetic peptides (BM17, BM36, BM40, BM100, BM101, and BM124) using mouse primary neurons.
4. Discussion In this study, we screened C-terminus-free peptide-displaying T7 phage libraries and successfully isolated a peptide sequence, BM17, as a lead for peptidic TrkB agonists. Because extension of the C-terminus or substitution of the C-terminal Phe16 by non-
aromatic residues resulted in remarkable reduction of its TrkBbinding affinity, BM17 would have strict constraints in its C-terminal structure. Therefore, if other display technologies such as filamentous M13 phage display, ribosome display, or mRNA display (which fuse to foreign peptides at the C-terminus) had been used, the BM17 sequence would not have been discovered.
Please cite this article in press as: T. Ohnishi, et al., Generation of a novel artificial TrkB agonist, BM17d99, using T7 phage-displayed random peptide libraries, Biochemical and Biophysical Research Communications (2017), http://dx.doi.org/10.1016/j.bbrc.2016.12.186
T. Ohnishi et al. / Biochemical and Biophysical Research Communications xxx (2017) 1e6 Table 2 Phage sequences obtained from the mutated BM17 library and their TrkB-binding activity. Clones
Sequence
Binding activity (fold)
BM17 BM17d99 BM17d156 BM17d123 BM17d144 BM17d169 BM17d43
NVRPRICRVRKWTLCF KSLPRMCRVRKWRLCF SVVSRMCRLRKYKLCF NVMQRICRIRNWRLCF TAIPRICGVRKWKLCF TGSLRICRVRKWKLCF KERPRVCRARNWRLCF
1.0 10.8 9.0 6.4 5.9 5.7 5.4
Binding activity indicates the fold-change value relative to the value for BM17, which was taken as 1.0.
Fig. 3. Affinity enhancement of BM17d99 peptide compared with parental BM17 peptide. Binding activity of biotinylated BM17 and BM17d99 peptides to TrkB-Fc. The binding activities were detected using horseradish peroxidase-conjugated streptavidin.
The BM17-displaying T7 phage exhibited direct TrkB agonistic activity in the cell-based assay, despite its lower concentration (Fig. 2A). Two key factors could be considered when obtaining such agonistic phages: (1) the multivalent peptide displayed on the phage, and (2) the self-dimerization property of the BM17 peptide. We speculated that the BM17-fused phage capsid coat proteins had self-dimerized during phage particle assembly in the host bacterial cells, and BM17 peptides were displayed on the outer surface of the T7 phage in a dimeric form. In fact, in a preliminary experiment, interactions between FAM-conjugated and TAMRA-conjugated BM17d99 peptides were observed, using the fluorescence resonance energy transfer phenomenon (data not shown). As a reported EPO-mimic peptide, Hematide also showed the selfdimerization property in crystal structure analysis [10]. This property might contribute to the agonistic activity of peptide ligands to
5
RTKs. The dimeric BM17d99(K1N/K11R) peptide exhibited partial TrkB phosphorylation activity (50e60% of full activity of BDNF) in the cell-based assay (Fig. 4). The interaction modes or configurations of TrkBeTrkB interaction (i.e., the distance of TrkB homodimerization) or ligandeTrkB interaction could be considered as the potential reason for the above results. Although the agonistic mechanism of BDNF and structure information on the TrkB/BDNF complex are still not fully known, the crystal structure of its related protein complex, TrkA/nerve growth factor, which has 50% sequence similarity to TrkB and BDNF, respectively, has been reported (PDB No. 1WWW) [17]. Based on this information, the distance of the farthest loops between two TrkA molecules is approximately 9 nm, and the distance of the closest loops between two TrkA molecules is approximately 2 nm, as calculated by the PyMOL molecular graphics system. Since the distance of the N-terminus between two peptides linked by bAsp is approximately 1 nm, it might be shorter for optical TrkB homodimerization. In the case of ligandeTrkB interaction, however, since the BM17 peptide did not bind TrkB in a BDNFcompetitive manner (data not shown), the binding mode of BM17 would be the other reason for its partial agonism. In fact, the TrkB agonistic antibodies reported by Wyeth Research and Pfizer Inc. [7,8]. bound TrkB in a BDNF-competitive manner and showed full agonistic activity. These points indicate that both TrkBeTrkB interaction and ligandeTrkB interaction would be needed for TrkB activation. Importantly, through a combination of phage panning against recombinant TrkB-Fc, direct agonistic screening of phage clones in cell-based assays, and use of chemical modification, we have successfully generated a novel artificial peptidic TrkB agonist. This approach is useful for generating artificial peptides that are agonistic to cell surface receptors. Author contributions T. Ohnishi conducted phage display screening and discovered peptide sequences. K. Sakamoto supervised phage display screening and wrote this paper. A. Shimizu conducted chemical synthesis of peptides. A. Asami-Odaka conducted in vitro cell-based assays. K. Nakamura conducted in vivo experiments and subsequent protein analyses. T. Ito conducted the recombinant protein preparations. T. Asami, T. Ohtaki, and H. Inooka supervised and supported this work. Conflict of interest No potential conflicts of interest are disclosed.
Fig. 4. TrkB agonistic activity of dimeric BM17d99(K1N/K11R) peptide. Phosphorylation of TrkB and Erk1/2 by the dimeric BM17d99(K1N/K11R) peptide, using TrkB-expressing HEK293 cells.
Please cite this article in press as: T. Ohnishi, et al., Generation of a novel artificial TrkB agonist, BM17d99, using T7 phage-displayed random peptide libraries, Biochemical and Biophysical Research Communications (2017), http://dx.doi.org/10.1016/j.bbrc.2016.12.186
6
T. Ohnishi et al. / Biochemical and Biophysical Research Communications xxx (2017) 1e6
Acknowledgements The authors would like to thank Drs. Yoshihiro Ishibashi, Kozo Hayashi, and Atsushi Nakanishi for help with the initial study of this work. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.bbrc.2016.12.186. References [1] G. Musumeci, L. Minichiello, BDNF-TrkB signalling in fear learning: from genetics to neural networks, Rev. Neurosci. 22 (2011) 303e315. [2] K. Martinowich, H. Manji, B. Lu, New insights into BDNF function in depression and anxiety, Nat. Neurosci. 10 (2007) 1089e1093. [3] A. Yoshii, M. Constantine-Paton, Postsynaptic BDNF-TrkB signaling in synapse maturation, plasticity, and disease, Dev. Neurobiol. 70 (2010) 304e322. [4] M. Yamanaka, Y. Itakura, T. Inoue, A. Tsuchida, T. Nakagawa, H. Noguchi, M. Taiji, Protective effect of brain-derived neurotrophic factor on pancreatic islets in obese diabetic mice, Metab. Clin. Exp. 55 (2006) 1286e1292. [5] S.M. Massa, T. Yang, Y. Xie, J. Shi, M. Bilgen, J.N. Joyce, D. Nehama, J. Rajadas, F.M. Longo, Small molecule BDNF mimetics activate TrkB signaling and prevent neuronal degeneration in rodents, J. Clin. Investig. 120 (2010) 1774e1785. [6] S.W. Jang, X. Liu, M. Yepes, K.R. Shepherd, G.W. Miller, Y. Liu, W.D. Wilson, G. Xiao, B. Blanchi, Y.E. Sun, K. Ye, A selective TrkB agonist with potent neurotrophic activities by 7,8-dihydroxyflavone, Proc. Nat. Acad. Sci. U. S. A. 107 (2010) 2687e2692. [7] M.D. Qian, J. Zhang, X.Y. Tan, A. Wood, D. Gill, S. Cho, Novel agonist monoclonal antibodies activate TrkB receptors and demonstrate potent neurotrophic activities, J. Neurosci. 26 (2006) 9394e9403.
[8] D. Tsa, H.K. Thomsen, J. Chou, J. Stratton, M. Hagen, C. Loo, C. Garcia, D.L. Sloane, A. Rosenthal, J.C. Lin, TrkB agonists ameliorate obesity and associated metabolic conditions in mice, Endocrinology 149 (2008) 1038e1048. [9] N.C. Wrighton, F.X. Farrell, R. Chang, A.K. Kashyap, F.P. Barbone, L.S. Mulcahy, D.L. Johnson, R.W. Barrett, L.K. Jolliffe, W.J. Dower, Small peptides as potent mimetics of the protein hormone erythropoietin, Science 273 (1996) 458e464. [10] O. Livnah, E.A. Stura, D.L. Johnson, S.A. Middleton, L.S. Mulcahy, N.C. Wrighton, W.J. Dower, J.K. Jolliffe, I.A. Wilson, Functional mimicry of a protein hormone by a peptide agonist: the EPO receptor complex at 2.8 Å, Science 273 (1996) 464e471. [11] M.W. Moyer, New blood-boosting drugs aim to staunch renal anemia, Nat. Med. 18 (2012) 332. [12] K. Sakamoto, Y. Ito, T. Mori, K. Sugimura, Interaction of human lactoferrin with cell adhesion molecules through RGD motif elucidated by lactoferrin-binding epitopes, J. Biol. Chem. 281 (2006) 24472e24478. [13] K. Sakamoto, Y. Ito, T. Hatanaka, P.B. Soni, T. Mori, K. Sugimura, Discovery and characterization of a peptide motif that specifically recognizes a non-native conformation of human IgG induced by acidic pH conditions, J. Biol. Chem. 284 (2009) 9986e9993. [14] S.E. Cwirla, P. Balasubramanian, D.J. Duffin, C.R. Wagstrom, C.M. Gates, S.C. Singer, A.M. Davis, R.L. Tansik, L.C. Mattheakis, C.M. Boytos, P.J. Schatz, D.P. Baccanari, N.C. Wrighton, R.W. Barrett, W.J. Dower, Peptide agonist of the thrombopoietin receptor as potent as the natural cytokine, Science 276 (1997) 1696e1699. [15] K. Ito, K. Sakai, Y. Suzuki, N. Ozawa, T. Hatta, T. Natsume, K. Matsumoto, H. Suga, Artificial human Met agonists based on macrocycle scaffolds, Nat. Commun. 6 (2009) 6375e6386. [16] K. Sakamoto, Y. Kawata, Y. Masuda, T. Umemoto, T. Ito, T. Asami, S. Takekawa, T. Ohtaki, H. Inooka, Discovery of an artificial peptide agonist to the fibroblast growth factor receptor 1c/bKlotho complex from random peptide T7 phage display, Biochem. Biophys. Res. Commun. 480 (2016) 55e60. [17] C. Wiesmann, M.H. Ultsch, S.H. Bass, A.M. de Vos, Crystal structure of nerve growth factor in complex with the ligand-binding domain of the TrkA receptor, Nature 401 (1999) 184e188.
Please cite this article in press as: T. Ohnishi, et al., Generation of a novel artificial TrkB agonist, BM17d99, using T7 phage-displayed random peptide libraries, Biochemical and Biophysical Research Communications (2017), http://dx.doi.org/10.1016/j.bbrc.2016.12.186
Contents lists available at ScienceDirect
Biochemical and Biophysical Research Communications journal homepage: www.elsevier.com/locate/ybbrc
Generation of a novel artificial TrkB agonist, BM17d99, using T7 phage-displayed random peptide libraries Toshiyuki Ohnishi 1, Kotaro Sakamoto*, 1, Asano Asami-Odaka, Kimie Nakamura, Ayako Shimizu, Takashi Ito, Taiji Asami, Tetsuya Ohtaki, Hiroshi Inooka Pharmaceutical Research Division, Takeda Pharmaceutical Company Limited, 26-1, Muraoka-Higashi 2-chome, Fujisawa, Kanagawa 251-8555, Japan
a r t i c l e i n f o
a b s t r a c t
Article history: Received 21 December 2016 Accepted 29 December 2016 Available online xxx
Tropomyosin receptor kinase B (TrkB) is a known receptor of brain-derived neurotrophic factor (BDNF). Because it plays a critical role in the regulation of neuronal development, maturation, survival, etc., TrkB is a good target for drugs against central nervous system diseases. In this study, we aimed to generate peptidic TrkB agonists by applying random peptide phage display technology. After the phage panning against recombinant Fc-fused TrkB (TrkB-Fc), agonistic phages were directly screened against TrkBexpressing HEK293 cells. Through subsequent screening of the first-hit BM17 peptide-derived focus library, we successfully obtained the BM17d99 peptide, which had no sequence similarity with BDNF but had TrkB-binding capacity. We then synthesized a dimeric BM17d99 analog peptide that could phosphorylate or activate TrkB by facilitating receptor homodimerization. Treatment of TrkB-expressing HEK293 cells with the dimeric BM17d99 analog peptide significantly induced the phosphorylation of TrkB, suggesting that homodimerization of TrkB was enhanced by the dimeric peptide. This report demonstrates that our approach is useful for the generation of artificial peptidic agonists of cell surface receptors. © 2016 Elsevier Inc. All rights reserved.
Keywords: Peptide Agonist TrkB BDNF Phage display
1. Introduction Tropomyosin receptor kinase B (TrkB) and its natural ligand, brain-derived neurotrophic factor (BDNF), which are expressed in many regions of the brain, are involved in psychiatric disorders (i.e., depression, schizophrenia) and neurodegenerative diseases (i.e., Alzheimer's disease, Huntington's disease) [1e3]. Despite these diverse pharmacological effects, BDNF has only a short half-life in the blood and poor central transition [4], making it difficult to use intrinsic BDNF as a therapeutic agent. Therefore, alternative molecules that mimic BDNF function would serve as novel innovative central nervous system (CNS) drugs. In recent years, small molecules, such as the LM22A series, have been reported as TrkB agonists [5]. However, these compounds did not show TrkB agonistic activity in our experimental system. The TrkB agonistic antibodies reported by Wyeth Research and Pfizer
Abbreviations: TrkB, tropomyosin-receptor-kinase B; BDNF, brain-derived neurotrophic factor; RTK, receptor tyrosine kinase. * Corresponding author. E-mail address: [email protected] (K. Sakamoto). 1 The first two authors contributed equally to this work.
Inc., respectively [6e8], showed remarkable specificity and agonistic activity, but they had no central transition ability because of their higher molecular weight. In contrast to these molecules, peptidic agonists of TrkB should possess binding specificity and the potential for CNS delivery. Since TrkB is a member of the receptor tyrosine kinases (RTKs), receptor dimerization is the key modification for its activation. The most successful example of an artificial peptidic agonist to RTK is the erythropoietin (EPO)-mimicking peptide Hematide [9e11], which was discovered through random peptide phage display technology and has no sequence similarity to EPO. Phage display is an extremely efficient method for the isolation of novel peptides that bind to target molecules [12,13]. Ever since the artificial peptidic agonist to RTKs was generated, several key achievements have been made: (1) discovery of a monomeric peptide with high affinity to the receptor, and (2) development of a potent dimerized peptide using a chemical linker, such as Lys (diamino acid), Asp (dicarboxylic acid), or bismaleimide. Although the process appears simple, it is actually quite difficult. Since the generation of Hematide, only a few artificial peptidic agonists to RTKs have been reported [14e16]. In this study, we aimed to obtain novel peptides that are
http://dx.doi.org/10.1016/j.bbrc.2016.12.186 0006-291X/© 2016 Elsevier Inc. All rights reserved.
Please cite this article in press as: T. Ohnishi, et al., Generation of a novel artificial TrkB agonist, BM17d99, using T7 phage-displayed random peptide libraries, Biochemical and Biophysical Research Communications (2017), http://dx.doi.org/10.1016/j.bbrc.2016.12.186
2
T. Ohnishi et al. / Biochemical and Biophysical Research Communications xxx (2017) 1e6
agonistic to TrkB. First, TrkB-Fc was constructed as a pseudoreceptor dimer for the affinity screening of T7 phage-displayed random peptide libraries. Next, the potential agonistic phages were directly screened against TrkB-expressing HEK293 cells. As a result, an agonistic phage (BM17) was successfully isolated, whose synthetic monomeric peptide exhibited significant agonistic activity in cell-based assays. Through affinity enhancement and chemical modification of the BM17 sequence, the potent TrkB agonist BM17d99 dimeric peptide was finally generated. 2. Materials and methods 2.1. Materials Protein constructs of the extracellular domain of human TrkB (Met1-His430) fused to the Fc portion of immunoglobulin G (TrkBFc) and human BDNF were produced internally, using a proprietary vector and expression in FreeStyle 293 cells. Human TrkA-Fc (1056TK), human TrkC-Fc (375-TC), and human p75NTR-Fc (367-NR) were purchased from R&D Systems (Minneapolis, MN, USA). A series of phage-derived peptides were chemically synthesized by employing Fmoc protection chemistry and were purified on a reversed phase column. Primary mouse cortical neurons were isolated from the cerebral cortex of E15 mice through use of a dissociation solution (MB-X0802; Sumitomo Bakelite Co., Ltd, Tokyo, Japan) and cultured in Neurobasal medium (21103049; Thermo Fisher Scientific, Waltham, MA, USA) containing 2% B27 supplement, GlutaMAX, and antibiotics. A human cell line stably expressing TrkB was constructed by transducing the human TrkB (variant C; GenBank No. NM_001018064)-expressing pcDNA3.1 vector to HEK293 cells and selecting for antibiotic resistance using geneticin (G418). These cells were maintained in Dulbecco's modified Eagle's medium (DMEM) containing 10% fetal bovine serum (FBS) and G418. 2.2. Phage library construction and panning The T7 phage-displayed random peptide libraries, which were generated by using mixed oligonucleotides as template DNA, were constructed by using a T7Select 10-3 vector from Merck (Darmstadt, Germany), according to methods described previously [12,13]. The recombinant Fc-fused protein was immobilized to Dynabeads Protein A/G (Invitrogen, Carlsbad, CA, USA) with 1.0% casein in phosphate-buffered saline (PBS). After washing with PBS containing 0.1% Tween 20 (PBST), the beads were incubated with the phage libraries for 1 h and then washed with PBST. The bound phages were eluted with 1% sodium dodecyl sulfate and then used to infect log-phase Escherichia coli BLT5615 cells (Merck) for phage amplification. After bacteriolysis, the phages were recovered from the culture supernatant by centrifugation and polyethylene glycol precipitation. The recovered phages were dissolved in PBS for use in the next round of panning. 2.3. Evaluation of phage- or synthetic peptide-binding to recombinant proteins by ELISA The wells of a Nunc Maxisorp microplate (460e518) were coated with anti-human Fc goat polyclonal antibody (109-005-098; Jackson ImmunoResearch, West Grove, PA, USA), followed by blocking with 1.0% casein in PBS. Fc-fused proteins were captured by the antibody, and the phage solution or biotinylated peptide solution was then added to the wells. After washing with PBST, the bound phages or peptides were detected by using horseradish peroxidase (HRP)-conjugated anti-T7 antibody (Merck) or HRP-conjugated streptavidin (Vector Laboratories, Burlingame, CA, USA),
respectively. The amounts of HRP in the wells were measured by the detection reagent tetramethylbenzidine (Wako Pure Chemical Industries, Osaka, Japan) or by a chemiluminescent reagent (Wako). 2.4. Chemical synthesis of dimeric peptide The peptide and triethylamine (12 equivalents) were dissolved in dimethylformamide (DMF), following which 0.1 M 4dimethylaminopyridine in DMF (0.1 equivalents) and the chemical linker Fmoc-bAsp(OSu)2 (bis(2,5-dioxopyrrolidin-1-yl) 3((((9H-fluoren-9-yl) methoxy) carbonyl) amino) pentanedioate) (M02722; Watanabe Chemical Instituted, Hiroshima, Japan) in DMF (0.5 equivalents) were added. The reaction mixture was mixed for 2 h, following which diethylamine was added to remove the Fmoc. After mixing for 15 min, 0.1 M acetic acid was added to terminate the reaction. The mixture was filtered through polytetrafluorethylene membrane filters, and purified by reversed phase highperformance liquid chromatography (Phenomenex C18 column, 4.6 4 150 mm). Pure fractions were collected and lyophilized to yield the dimeric peptide as a white powder. 2.5. Evaluation of agonistic activity of phages or synthetic peptides in cell-based assay Human TrkB-expressing HEK293 cells were plated on poly-DLys-coated plates in DMEM containing 10% FBS and G418. After exchanging the medium to 0.1% BSA-containing DMEM, phages or synthetic peptides diluted with 0.05% CHAPS-containing DMEM were added to the wells and allowed to incubate for 30 min. Then, the phage solution or peptide solution was removed, and the cells were lysed with RIPA buffer (pH 7.5; 50 mM Tris-HCl, 1%Triton X100, 150 mM NaCl, and 5 mM ethylenediaminetetraacetic acid) containing a protease inhibitor cocktail and phosphatase inhibitors (Roche, Basel, Switzerland). The cell lysates were electrophoresed and then transferred to a polyvinylidene fluoride membrane by the TransBlot Turbo system (Bio-Rad, Hercules, CA, USA.). For detection of phosphorylated TrkB or extracellular signal-regulated kinases 1 and 2 (Erk1/2), anti-pY516 rabbit monoclonal antibody (Cell Signaling Technology, Beverly, MA, USA) or anti-p44/42 MAPK (T202/Y204) rabbit polyclonal antibody (Cell Signaling Technology) were used, respectively. For the secondary antibody, IRDye 800CWlabeled anti-rabbit IgG antibody (1:800; #611-131-122; Rockland Immunochemicals Inc., Limerick, PA, USA) was used, and the band intensity was measured at the extension and emission wavelengths of 780 and 820 nm, respectively, with the use of the Odyssey Imaging System (Li-COR, Lincoln, NE, USA). When performing the experiments with primary mouse cortical neurons, Neurobasalbased culture medium was used. 3. Results 3.1. Recombinant TrkB-Fc binding phages were isolated from T7 phage libraries To obtain TrkB-binding peptides, random peptide-displaying T7 phage libraries were panned against TrkB-Fc. At the 5th round of panning, TrkB-binding polyclonal phages were concentrated (Fig. 1A), and a total of 475 clones were picked to test their TrkBbinding activity. In combination with sequence analysis, 11 independent sequences having TrkB-binding affinity were obtained (Table 1). These clones shared consensus motifs with one another and had no sequence similarity with BDNF. Although these 11 phage clones bound selectively to TrkB-Fc (the results from representative phage clone BM17 are shown in Fig. 1B), they had no competitiveness with BDNF in TrkB binding (data not shown).
Please cite this article in press as: T. Ohnishi, et al., Generation of a novel artificial TrkB agonist, BM17d99, using T7 phage-displayed random peptide libraries, Biochemical and Biophysical Research Communications (2017), http://dx.doi.org/10.1016/j.bbrc.2016.12.186
T. Ohnishi et al. / Biochemical and Biophysical Research Communications xxx (2017) 1e6
3
Fig. 1. Biopanning of random peptide-displaying phage libraries against TrkB-Fc. (A) Binding activity of polyclonal phages to TrkB-Fc in ELISA. The Fc-fusion proteins were immobilized by anti-Fc antibody, and the phage binding was detected with horseradish peroxidase-conjugated anti-T7 phage antibody. (B) Binding specificity of monoclonal phage BM17. The phage binding activity to TrkB-Fc, TrkA-Fc, TrkC-Fc, and p75NTR-Fc was evaluated. The number of phages was expressed as plaque-forming units (pfu) per well.
Table 1 TrkB-binding sequences obtained from T7 phage display. Cluster
Name
Sequence
1 1 1
BM17 BM40 BM36
NVRPRICRVRKWTLCF PAKHLCLLAANRVRRSWQCL PRHKKCFRVLRACIEY
2 2 2
BM97 BM124 BM101
SDTCWKFYRGWPRRCRSN HAWCFPGPSLWPRRCRNN KSWCRQGFWPIRCSTT
3 3
BM15 BM64
GPWKLSFL GPWNYRSYKFWSKTAK
Singleton Singleton Singleton
BM07 BM74 BM100
AWPYQNWSFWTSRFSD NIPSIRDCIPIGCLQYAFWR KRACKHSHGIWCTKW
panning against TrkB-Fc, sequence analysis revealed that the phages with no TrkB-binding activity had a Phe16 deletion or Cterminal elongation (data not shown). Based on the above results, another BM17-derived T7 phage library was reconstructed, which displayed peptides having several mutations within the BM17 sequence without changing the chain length. The screening showed several phage clones with increased TrkB-binding activity (Table 2). In particular, phage BM17d99 (KSLPRMCRVRKWRLCF) had the highest affinity, being 10.8-times higher than that of the parental BM17 phage (Table 2). The EC50 binding values for the BM17 and BM17d99 synthetic peptides were 17 nM and 0.45 nM, respectively (Fig. 3), demonstrating BM17d99 to have a 38-times higher binding affinity than the parental BM17 peptide. The relative affinity between the BM17 and BM17d99 sequences was consistent in both peptide and phage forms.
Underlines indicate conserved or similar amino acid positions.
3.4. Dimeric BM17d99 peptide analog activated cell surface TrkB 3.2. BM17 sequence showed agonistic activity both in phage form and as a synthetic peptide Next, we directly evaluated the agonistic activity of the 11 phage clones in a cell-based assay. Three (BM17, BM100, and BM124) out of the 11 clones induced phosphorylation of both TrkB and its downstream Erk1/2 (Fig. 2A), even though the concentration of the particular peptide displayed in 2.5 1011 pfu/mL of phage suspension was very low (approximately 420 pM, assuming one phage particle corresponds to one peptide molecule). By additionally selecting two clones (BM36 and BM40) that have sequence similarity to BM17 to examine the structure-activity relationships, a total of 5 sequences (BM17, BM36, BM40, BM100, and BM124) were chemically synthesized. Among these, only peptide BM17 (NVRPRICRVRKWTLCF) exhibited significant phosphorylation of TrkB in mouse primary neurons (Fig. 2B). 3.3. BM17d99 sequence was 10-times more potent than the parental BM17 sequence For affinity enhancement, a BM17-derived T7 phage library (NVRPRICRVRKWTLCXXXXX) was constructed, where 5 random amino acid residues were added at the C-terminus of BM17. After
As described above, the monomeric BM17 peptide induced phosphorylation of TrkB, even though its agonistic activity was very weak (EC50 ¼ 104 M). Since receptor homodimerization is needed for TrkB activation, we hypothesized that dimerization of the BM17d99 peptide, by conjugation of its N-terminus to the dicarboxylic acid-linker bAsp, might increase its agonistic activity. Prior to examining this hypothesis, Lys1 and Lys11, both with reactive amino groups in their side chains, were substituted to Asn and Arg, respectively, to avoid undesired conjugate reactions. The resultant monomeric BM17d99(K1N/K11R) peptide showed an approximately 5-fold reduction in TrkB-Fc-binding activity in comparison with BM17d99, as demonstrated by ELISA (data not shown). However, the conjugated dimeric BM17d99(K1N/K11R) peptide demonstrated an almost 10,000-fold increase in its agonistic activity (EC50 ¼ 10 nM) compared with that of the monomeric BM17 peptide, suggesting that peptide dimerization by linkage had successfully led to the cooperative binding of the BM17d99(K1N/K11R) moieties to the TrkB monomers, with subsequent TrkB homodimerization (Fig. 4). Moreover, intracerebroventricular administration of the dimeric BM17d99(K1N/K11R) peptide to C57BL/6NCrl mice significantly induced TrkB phosphorylation in the hypothalamus, suggesting that our peptide could activate cell surface TrkB not only in vitro but also in vivo (Supplementary Fig. S1).
Please cite this article in press as: T. Ohnishi, et al., Generation of a novel artificial TrkB agonist, BM17d99, using T7 phage-displayed random peptide libraries, Biochemical and Biophysical Research Communications (2017), http://dx.doi.org/10.1016/j.bbrc.2016.12.186
4
T. Ohnishi et al. / Biochemical and Biophysical Research Communications xxx (2017) 1e6
Fig. 2. Evaluation of TrkB-agonistic activity of phages and synthetic peptides. (A) Phosphorylation of TrkB and Erk1/2 by phage particles (2.5 1011 pfu/mL) using TrkBexpressing HEK293 cells. (B) Phosphorylation of TrkB by synthetic peptides (BM17, BM36, BM40, BM100, BM101, and BM124) using mouse primary neurons.
4. Discussion In this study, we screened C-terminus-free peptide-displaying T7 phage libraries and successfully isolated a peptide sequence, BM17, as a lead for peptidic TrkB agonists. Because extension of the C-terminus or substitution of the C-terminal Phe16 by non-
aromatic residues resulted in remarkable reduction of its TrkBbinding affinity, BM17 would have strict constraints in its C-terminal structure. Therefore, if other display technologies such as filamentous M13 phage display, ribosome display, or mRNA display (which fuse to foreign peptides at the C-terminus) had been used, the BM17 sequence would not have been discovered.
Please cite this article in press as: T. Ohnishi, et al., Generation of a novel artificial TrkB agonist, BM17d99, using T7 phage-displayed random peptide libraries, Biochemical and Biophysical Research Communications (2017), http://dx.doi.org/10.1016/j.bbrc.2016.12.186
T. Ohnishi et al. / Biochemical and Biophysical Research Communications xxx (2017) 1e6 Table 2 Phage sequences obtained from the mutated BM17 library and their TrkB-binding activity. Clones
Sequence
Binding activity (fold)
BM17 BM17d99 BM17d156 BM17d123 BM17d144 BM17d169 BM17d43
NVRPRICRVRKWTLCF KSLPRMCRVRKWRLCF SVVSRMCRLRKYKLCF NVMQRICRIRNWRLCF TAIPRICGVRKWKLCF TGSLRICRVRKWKLCF KERPRVCRARNWRLCF
1.0 10.8 9.0 6.4 5.9 5.7 5.4
Binding activity indicates the fold-change value relative to the value for BM17, which was taken as 1.0.
Fig. 3. Affinity enhancement of BM17d99 peptide compared with parental BM17 peptide. Binding activity of biotinylated BM17 and BM17d99 peptides to TrkB-Fc. The binding activities were detected using horseradish peroxidase-conjugated streptavidin.
The BM17-displaying T7 phage exhibited direct TrkB agonistic activity in the cell-based assay, despite its lower concentration (Fig. 2A). Two key factors could be considered when obtaining such agonistic phages: (1) the multivalent peptide displayed on the phage, and (2) the self-dimerization property of the BM17 peptide. We speculated that the BM17-fused phage capsid coat proteins had self-dimerized during phage particle assembly in the host bacterial cells, and BM17 peptides were displayed on the outer surface of the T7 phage in a dimeric form. In fact, in a preliminary experiment, interactions between FAM-conjugated and TAMRA-conjugated BM17d99 peptides were observed, using the fluorescence resonance energy transfer phenomenon (data not shown). As a reported EPO-mimic peptide, Hematide also showed the selfdimerization property in crystal structure analysis [10]. This property might contribute to the agonistic activity of peptide ligands to
5
RTKs. The dimeric BM17d99(K1N/K11R) peptide exhibited partial TrkB phosphorylation activity (50e60% of full activity of BDNF) in the cell-based assay (Fig. 4). The interaction modes or configurations of TrkBeTrkB interaction (i.e., the distance of TrkB homodimerization) or ligandeTrkB interaction could be considered as the potential reason for the above results. Although the agonistic mechanism of BDNF and structure information on the TrkB/BDNF complex are still not fully known, the crystal structure of its related protein complex, TrkA/nerve growth factor, which has 50% sequence similarity to TrkB and BDNF, respectively, has been reported (PDB No. 1WWW) [17]. Based on this information, the distance of the farthest loops between two TrkA molecules is approximately 9 nm, and the distance of the closest loops between two TrkA molecules is approximately 2 nm, as calculated by the PyMOL molecular graphics system. Since the distance of the N-terminus between two peptides linked by bAsp is approximately 1 nm, it might be shorter for optical TrkB homodimerization. In the case of ligandeTrkB interaction, however, since the BM17 peptide did not bind TrkB in a BDNFcompetitive manner (data not shown), the binding mode of BM17 would be the other reason for its partial agonism. In fact, the TrkB agonistic antibodies reported by Wyeth Research and Pfizer Inc. [7,8]. bound TrkB in a BDNF-competitive manner and showed full agonistic activity. These points indicate that both TrkBeTrkB interaction and ligandeTrkB interaction would be needed for TrkB activation. Importantly, through a combination of phage panning against recombinant TrkB-Fc, direct agonistic screening of phage clones in cell-based assays, and use of chemical modification, we have successfully generated a novel artificial peptidic TrkB agonist. This approach is useful for generating artificial peptides that are agonistic to cell surface receptors. Author contributions T. Ohnishi conducted phage display screening and discovered peptide sequences. K. Sakamoto supervised phage display screening and wrote this paper. A. Shimizu conducted chemical synthesis of peptides. A. Asami-Odaka conducted in vitro cell-based assays. K. Nakamura conducted in vivo experiments and subsequent protein analyses. T. Ito conducted the recombinant protein preparations. T. Asami, T. Ohtaki, and H. Inooka supervised and supported this work. Conflict of interest No potential conflicts of interest are disclosed.
Fig. 4. TrkB agonistic activity of dimeric BM17d99(K1N/K11R) peptide. Phosphorylation of TrkB and Erk1/2 by the dimeric BM17d99(K1N/K11R) peptide, using TrkB-expressing HEK293 cells.
Please cite this article in press as: T. Ohnishi, et al., Generation of a novel artificial TrkB agonist, BM17d99, using T7 phage-displayed random peptide libraries, Biochemical and Biophysical Research Communications (2017), http://dx.doi.org/10.1016/j.bbrc.2016.12.186
6
T. Ohnishi et al. / Biochemical and Biophysical Research Communications xxx (2017) 1e6
Acknowledgements The authors would like to thank Drs. Yoshihiro Ishibashi, Kozo Hayashi, and Atsushi Nakanishi for help with the initial study of this work. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.bbrc.2016.12.186. References [1] G. Musumeci, L. Minichiello, BDNF-TrkB signalling in fear learning: from genetics to neural networks, Rev. Neurosci. 22 (2011) 303e315. [2] K. Martinowich, H. Manji, B. Lu, New insights into BDNF function in depression and anxiety, Nat. Neurosci. 10 (2007) 1089e1093. [3] A. Yoshii, M. Constantine-Paton, Postsynaptic BDNF-TrkB signaling in synapse maturation, plasticity, and disease, Dev. Neurobiol. 70 (2010) 304e322. [4] M. Yamanaka, Y. Itakura, T. Inoue, A. Tsuchida, T. Nakagawa, H. Noguchi, M. Taiji, Protective effect of brain-derived neurotrophic factor on pancreatic islets in obese diabetic mice, Metab. Clin. Exp. 55 (2006) 1286e1292. [5] S.M. Massa, T. Yang, Y. Xie, J. Shi, M. Bilgen, J.N. Joyce, D. Nehama, J. Rajadas, F.M. Longo, Small molecule BDNF mimetics activate TrkB signaling and prevent neuronal degeneration in rodents, J. Clin. Investig. 120 (2010) 1774e1785. [6] S.W. Jang, X. Liu, M. Yepes, K.R. Shepherd, G.W. Miller, Y. Liu, W.D. Wilson, G. Xiao, B. Blanchi, Y.E. Sun, K. Ye, A selective TrkB agonist with potent neurotrophic activities by 7,8-dihydroxyflavone, Proc. Nat. Acad. Sci. U. S. A. 107 (2010) 2687e2692. [7] M.D. Qian, J. Zhang, X.Y. Tan, A. Wood, D. Gill, S. Cho, Novel agonist monoclonal antibodies activate TrkB receptors and demonstrate potent neurotrophic activities, J. Neurosci. 26 (2006) 9394e9403.
[8] D. Tsa, H.K. Thomsen, J. Chou, J. Stratton, M. Hagen, C. Loo, C. Garcia, D.L. Sloane, A. Rosenthal, J.C. Lin, TrkB agonists ameliorate obesity and associated metabolic conditions in mice, Endocrinology 149 (2008) 1038e1048. [9] N.C. Wrighton, F.X. Farrell, R. Chang, A.K. Kashyap, F.P. Barbone, L.S. Mulcahy, D.L. Johnson, R.W. Barrett, L.K. Jolliffe, W.J. Dower, Small peptides as potent mimetics of the protein hormone erythropoietin, Science 273 (1996) 458e464. [10] O. Livnah, E.A. Stura, D.L. Johnson, S.A. Middleton, L.S. Mulcahy, N.C. Wrighton, W.J. Dower, J.K. Jolliffe, I.A. Wilson, Functional mimicry of a protein hormone by a peptide agonist: the EPO receptor complex at 2.8 Å, Science 273 (1996) 464e471. [11] M.W. Moyer, New blood-boosting drugs aim to staunch renal anemia, Nat. Med. 18 (2012) 332. [12] K. Sakamoto, Y. Ito, T. Mori, K. Sugimura, Interaction of human lactoferrin with cell adhesion molecules through RGD motif elucidated by lactoferrin-binding epitopes, J. Biol. Chem. 281 (2006) 24472e24478. [13] K. Sakamoto, Y. Ito, T. Hatanaka, P.B. Soni, T. Mori, K. Sugimura, Discovery and characterization of a peptide motif that specifically recognizes a non-native conformation of human IgG induced by acidic pH conditions, J. Biol. Chem. 284 (2009) 9986e9993. [14] S.E. Cwirla, P. Balasubramanian, D.J. Duffin, C.R. Wagstrom, C.M. Gates, S.C. Singer, A.M. Davis, R.L. Tansik, L.C. Mattheakis, C.M. Boytos, P.J. Schatz, D.P. Baccanari, N.C. Wrighton, R.W. Barrett, W.J. Dower, Peptide agonist of the thrombopoietin receptor as potent as the natural cytokine, Science 276 (1997) 1696e1699. [15] K. Ito, K. Sakai, Y. Suzuki, N. Ozawa, T. Hatta, T. Natsume, K. Matsumoto, H. Suga, Artificial human Met agonists based on macrocycle scaffolds, Nat. Commun. 6 (2009) 6375e6386. [16] K. Sakamoto, Y. Kawata, Y. Masuda, T. Umemoto, T. Ito, T. Asami, S. Takekawa, T. Ohtaki, H. Inooka, Discovery of an artificial peptide agonist to the fibroblast growth factor receptor 1c/bKlotho complex from random peptide T7 phage display, Biochem. Biophys. Res. Commun. 480 (2016) 55e60. [17] C. Wiesmann, M.H. Ultsch, S.H. Bass, A.M. de Vos, Crystal structure of nerve growth factor in complex with the ligand-binding domain of the TrkA receptor, Nature 401 (1999) 184e188.
Please cite this article in press as: T. Ohnishi, et al., Generation of a novel artificial TrkB agonist, BM17d99, using T7 phage-displayed random peptide libraries, Biochemical and Biophysical Research Communications (2017), http://dx.doi.org/10.1016/j.bbrc.2016.12.186