An Orally Active Bradykinin B1 Receptor Antagonist Engineered as a

Dec 5, 2016 - An orally active and metabolically stable peptide TIBA was successfully engineered as a chimera by fusing an analgesic bradykinin recept...
0 downloads 8 Views 808KB Size
Subscriber access provided by The University of Liverpool

Brief Article

An orally active bradykinin B1 receptor antagonist engineered as a bifunctional chimera of sunflower trypsin inhibitor Yibo Qiu, Misako Taichi, Na Wei, Huan Yang, Kathy Qian Luo, and James P. Tam J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.6b01011 • Publication Date (Web): 05 Dec 2016 Downloaded from http://pubs.acs.org on December 7, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Journal of Medicinal Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 7

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Medicinal Chemistry

An orally active bradykinin B1 receptor antagonist engineered as a bifunctional chimera of sunflower trypsin inhibitor Yibo Qiu,1 Misako Taichi,2 Na Wei,3 Huan Yang,4 Kathy Qian Luo,5 and James P. Tam*1 1School of Biological Sciences, Nanyang Technological University, 60 Nanyang Drive, 637551, Singapore 2Biofunctional Synthetic Chemistry Laboratory, RIKEN 2-1 Hirosawa, Wako-shi, Saitama, 351-0198, Japan 3School of Chemical and Biomedical Engineering, Nanyang Technological University, 62 Nanyang Drive, 637459, Singapore 4School of Pharmacy, Jiangsu University, 301 Xuefu Road, Zhenjiang, Jiangsu, 212013, P. R. China 5Faculty of Health Sciences, University of Macau, Taipa, Macau, P. R. China ABSTRACT: An orally active and metabolically stable peptide TIBA was successfully engineered as a chimera by fusing an analgesic bradykinin receptor antagonist peptide and the trypsin inhibitory loop of sunflower trypsin inhibitor-1. As a fusion cyclic peptide, the metabolically labile analgesic peptide is protected from degradation by exopeptidases as well as the endopeptidases, and its serum half-life extended from 6 h as a chimera. Moreover, the chimera TIBA was also found to be orally active in an animal pain model using a hot plate assay.

■INTRODUCTION Bradykinin, a positively charged nonapeptide and neuroactive hormone that mediates pain, inflammation, and vasodi1 lation, is the most potent elicitor of pain, and its role in no2 ciceptive response has been studied over the past fifty years. The bradykinin-mediated pain process involves two types of G-protein-coupled receptors, namely B1 and B2 receptors. While B2 receptors are constitutively expressed, B1 receptors are expressed at a low level in healthy tissues and induced 3, 4 only by tissue injury or noxious stimulation. As such, bradykinin B1 receptors that exhibit disease-dependent expression would be more suitable drug targets for pain management than B2 receptors. A large number of B1 receptor antagonists have been reported including arylsulfonamide- and amide-based small molecule antagonists, of which some candidates reached 5 phase II clinical trials. Among the different types of antagonists developed thus far, the peptidyl bradykinin B1 receptor antagonists with their large footprints are shown to be po6 tent and specific. However, two limitations have hampered their development as therapeutics. They are metabolically labile, with a half-life of 95% purity and analyzed by MS (Figure S2-7). Comparison of the Hα chemical shifts of TIBA, SLBA and SFTI-1 showed that, except for the DALK region, TIBA and SLBA were highly similar to SFTI-1 (Figure S8). Although TIBA and SLBA are larger than the scaffold SFTI-1, the chemical shifts of TI loop in TIBA and SL loop in SLBA overlap nicely with corresponding parts in the scaffold. Scheme 2. A scheme demonstrating an amide-to-amide cyclization. PG: protecting groups; Trt: trityl protecting group.

Table 1. Sequences of SFTI-1, DALK and grafted peptide analogs.

For cyclic peptides, we employed an amide-to-amide cy28, 29 clization strategy as shown in Scheme 2. This scheme uses an amide linkage to the resin support during a solidphase peptide assemblage stage and this amide linkage is

Proteins and peptides are generally unstable during heat, acid hydrolysis and proteolytic enzyme treatment, resulting in loss of their bioactivity as therapeutics. To test the stability of our peptide analogs under heat, acid or proteolytic

ACS Paragon Plus Environment

Page 3 of 7

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Medicinal Chemistry

treatments, they were boiled at 100 °C, treated with hydrochloride solution or incubated with proteases and their integrity determined by HPLC and MS. Overall, both cyclic and linear peptides were found to be relatively stable under heat and acidic conditions as they retained >80% integrity after 6 h, suggesting that peptides tested here would likely sustain the heating process and survive the acidic environment of the stomach. (Figure S9). In proteolytic digestion assays, the trypsin inhibitory (TI)loop-containing chimera TIBA was stable under both pepsin and trypsin treatment whereas the SLBA without the TI loop was only stable against pepsin but completely digested by trypsin in 6 h (Figure 1A and S9), suggesting that the TI loop improved stability against trypsin. In comparison, both linear counterparts, TIBA-L and SLBA-L, were relatively unstable, underscoring the importance of a cyclic structure. Overall, the DALK peptide was degraded rapidly, with >50% being digested by pepsin in 6 h. It should be noted that DALK is 1 resistant to trypsin digestion because Lys is located at the Nterminus which is not recognized by the endopeptidase tryp2 sin and cleavage at Arg is blocked by the Pro following it. Collectively, our results showed our design strategy as a cyclic chimera with a built-in TI-loop greatly improved the stability and resistance to enzymatic digestion of DALK.

Figure 1. Trypsin and serum stability of engineered peptides. Peptide stability of DALK (●), SLBA (■), SLBA-L (▲), TIBA (▼), TIBA-L (♦) and SFTI-1 (●) under A) trypsin digestion, B) serum treatment. The Y axis represents the relative amount of peptides remaining at each time points compared to the starting time point (0 h). Experiments are repeated in triplicates and results are shown as mean ± SD. Previous studies have shown that bradykinin was rapidly degraded into a five-residue peptide in human serum by aminopeptidase P, dipeptidyl peptidase IV, and an angioten32 sin-converting enzyme. Considering the high sequence similarity between DALK and bradykinin, DALK was likely to be digested by these serum enzymes. Stability assay also showed that both TIBA and SLBA, were fairly stable in human serum up to 6 h, with >90% remaining intact, whereas DALK was largely degraded in 2 h (Figure 1B). The current 7 results are in agreement with our previous work. Taken together, these results strongly suggest that through engineering, the serum stability of DALK was remarkably improved. The high serum stability of TIBA provides a premise for its therapeutic use in vivo. In our trial in vivo experiment using the dose of 10 mg/kg by oral administration, TIBA was detected in serum at a concentration of 11.52 µM after 4 h (data not shown). The trypsin inhibitory activity of SFTI-1 largely stems from 33 5 6 its trypsin inhibitory loop. The Lys -Ser in this loop serves as a substrate of trypsin but can be regenerated by a trypsinmediated ligation. As a result, SFTI-1 efficiently inhibits the catalytic activity of trypsin by occupying the active site. Since

TIBA contains this loop, we proceeded to investigate whether it retained the inhibitory properties of the scaffold SFTI-1. The trypsin inhibition assays were performed using a α chromogenic trypsin substrate N -benzoyl-DL-arginine 4nitroanilide hydrochloride at 37°C. The activity of trypsin was determined by measuring the absorbance at 405 nm. TIBA exhibited a similar inhibitory effect as SFTI-1, whereas no inhibitory activity was observed for SLBA at concentrations up to 100 µM (Figure S10). These results confirmed that TIBA, which contains the trypsin inhibitory loop, retained the trypsin inhibition activity as SFTI-1. In contrast, SLBA, which does not contain the trypsin inhibitory loop, was completely inactive, suggesting that the trypsin inhibition activity is dependent on the trypsin inhibitory loop. Notably, TIBA-L, which contains a trypsin inhibitory loop but not a cyclic backbone, was a poor trypsin inhibitor, suggesting the importance of a cyclic backbone in maintaining the secondary structure for trypsin binding. We further determined the inhibition constant of TIBA to be 5.12 nM which was similar as SFTI-1 of 2.35 nM (Figure 2). Our results are within the range of reported inhibition constants of SFTI-1 by other 21 34 35 groups (0.1 nM, 1 nM, and 13 nM ).

Figure 2. Dixon plot for the determination of the dissociation constant (Ki) of SFTI-1 and TIBA at 2 mM (●) and 3 mM (♦) BAPNA. The Y axis represents the reciprocal of the velocity and plotted against different concentrations of SFTI-1 or TIBA. Experiments are repeated in triplicates and results are shown as mean ± SD. The equations for SFTI-1 are Y = 6.368*X + 100.05 (R² = 0.9985) and Y = 3.364*X + 92.98 (R² =0.9939). The equations for TIBA are Y = 5.778*X + 103.03 (R² = 0.9948) and Y = 1.728*X + 82.26 (R² = 0.9985). To determine the bradykinin B1 receptor antagonist activi2+ ties of TIBA and SLBA, an intracellular Ca assay was performed. Bradykinin B1 receptor activation leads to activation of phospholipase C and the release of inositol phosphates, 2+ thereby triggering an increase in the intracellular Ca level. A bradykinin B1 receptor antagonist would suppress cellular 2+ activation, resulting in a decrease in Ca levels. In this assay, TM cells were pre-loaded with a fluorescent dye (fluo-4 Direct ) 2+ to determine the intracellular Ca level. Before performing 2+ the Ca assay, we determined that our chimeric peptides were not cytotoxic to HeLa cells at concentrations up to 40 µM (Figure S11). Cells were first treated with the chimeric peptides at concentrations up to 10 µM. Measurements were then performed after adding bradykinin B1 receptor agonist 9 2+ des-Arg -BK (100 nM) to induce Ca . Figure 3 showed that TIBA, SLBA and DALK were all able to suppress the increase 2+ in Ca concentrations at micromolar concentrations. The IC50 of TIBA, SLBA and DALK (positive control) was 2.00±0.33 µM, 4.04±0.64 μM and 0.99±0.42 μM, respectively (Table 2). This is also partly supported by our NMR data (Figure S12). Ha et al. showed that a turned conformation is necessary for antagonist to adopt the active conformation and then interact with the binding site of bradykinin B1 re-

ACS Paragon Plus Environment

Journal of Medicinal Chemistry 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

4

5

ceptors. As shown in Figure S12, the Pro -Pro dipeptide in the DALK region of TIBA (indicated by the arrow) did exist in a turned structure which is critical for the receptor binding. It should be noted that the IC50 of DALK in this paper is substantially higher than those reported by Hawkinson et al 37 2+ (µM vs nM). This is because in their Ca assay, a sensitizing step was included to increase the expression level of B1 receptor, and which was not performed in our case. B1 receptors are normally expressed at a low level in most tissues, but will be up regulated through inflammation. Thus, a low IC50 value of DALK could be achieved in vitro with a sensitization step. In addition, both cell types (HeLa vs IMR-90 human lung fibroblast cells) and agonist concentrations (100 nM vs 4 to 10 nM) were different to contribute to observed results.

Figure 3. Bradykinin receptor antagonism activities of SFTI-1 (▲), SLBA (●), TIBA (■) and DALK (♦). The Y axis represents the fluorescence change in the HeLa cells, which can be used 2+ to indicate the intracellular Ca level. Peptides were tested at a concentration from 0 to 10 μM. The increase of the fluorescence was due to the pretreatment of a B1 agonist (RPPGFSPF) with HeLa cells. Peptides with an antagonism activity against bradykinin B1 receptor would result in a decrease of the fluorescence. 2+

Table 2. IC50 for the inhibitory effects of Ca the presence of engineered peptides

release in

(IC50, concentrations resulting in 50% of growth inhibition; 95% CI, 95% confidence intervals.) To evaluate whether TIBA is orally active in vivo, a hot plate assay was conducted in mice using both intraperitoneal injection (i.p.) and oral administration (p.o.). B1 receptors are not only related to chronical pain but also play an important role in spinal cord plasticity modulation, which underlies the central component of pain. The hot plate assay, different from other assays using noxious heat stimulus, is a behaviorbased assay strongly modulated at spinal and supraspinal levels, and thus can be used for evaluation of B1 receptor38 induced pain. The pain sensation caused by thermal stimuli should evoke a rapid hind paw-licking reaction in mice, which indicates a nociceptive response to the elevated temperature. The licking of the front paws is seen as a grooming

Page 4 of 7

response for mice and not necessarily connected to an un39 pleasant pain sensation, thus it was not considered as a pain response in this study. The time it takes for a mouse to lick its hind paw from the time it is put on the hot plate was defined as the latency. The baseline latency and the latency after receiving the peptide drugs were compared, and an increment in the response latency represented an analgesic effect. Our results showed that TIBA was able to relieve pain via both i.p. and p.o. administration, but DALK was only active through i.p. injection (Figure 4A and B). An i.p. injection of either DALK or TIBA was able to increase the pain response latency. They inhibited the pain response compared to mice injected with phosphate-buffered saline and increased the response latency by 201% and 125%, respectively, at the maxi-

Figure 4. Analgesic effects of DALK, TIBA and SFTI-1 in mice models via i.p. and p.o. administrations. Hot plate assay was performed on DALK (●), TIBA (■) and SFTI-1 (▲) at different concentrations via A) intraperitoneal (i.p.) and B) oral (p.o.) administrations. The Y axis represents percentages of increment on mice’s response time towards thermal induced pain. The X axis represents concentrations used for each peptide. The data are shown as mean ± S.E.M. (n=6-8) (* p