This review deals with short peptides (up to 50 amino acids) as biomimetic active recognition elements in sensing systems. Peptide-based sensors have been developed in recent years according to different strategies. Synthetic peptides have been designed on the basis of known interactions between single or a few amino acids and targets, with attention being paid to the presence of peptide motifs known to allow intermolecular self-organization of the sensing peptides over the sensor surface. Sensitive and sophisticated sensors have been obtained in this way, but the use of designed peptides is limited by severe difficulties in their in silico design. Short peptides from random phage display have been selected in a random way from large, unfocussed, and often preexisting and commercially available phage display libraries, with no design elements. Such peptides often perform better than antibodies, but they are difficult to select when the target is a small molecule because of the need to immobilize it with considerable modifications of its structure. Artificial, miniaturized receptors have been obtained from the reduction of the known sequence of a natural receptor down to a synthesizable and yet stable one. Alternatively, binding sites have been created over a designed, stable peptide scaffold. Short peptides have also been used as active elements for the detection of their own natural receptors: pathogenic bacteria have been detected with antimicrobial and cell-penetrating peptides, but key challenges such as detection of bacteria in real samples, improved sensitivity, and improved selectivity have to be faced. Peptide substrates have been conjugated to fluorescent quantum dots to obtain disposable sensors for protease activity with high sensitivity. Ferrocene-peptide conjugates have been used for electrochemical sensing of protease activity.

Bicycles are constrained bicyclic peptides that represent a promising binding modality for use in targeted drug conjugates. A phage display screen against EphA2, a receptor tyrosine kinase highly expressed in a number of solid tumors, identified a number of Bicycle families with low nanomolar affinity. A Bicycle toxin conjugate (BTC) was generated by derivatization of one of these Bicycles with the potent cytotoxin DM1 via a cleavable linker. This BTC demonstrated potent antitumor activity in vivo but was poorly tolerated, which was hypothesized to be the result of undesired liver uptake caused by poor physicochemical properties. Chemical optimization of a second Bicycle, guided by structural biology, provided a high affinity, metabolically stable Bicycle with improved physicochemical properties. A BTC incorporating this Bicycle also demonstrated potent antitumor activity and was very well tolerated when compared to the initial BTC. Phage display selection followed by chemical optimization of Bicycles can deliver potent drug conjugates with favorable pharmaceutical properties.

The EphA2 receptor is found at high levels in tumors and low levels in normal tissue and high EphA2 expression in biopsies is a predictor of poor outcome in patients. Drug discovery groups have therefore sought to develop EphA2-based therapies using small molecule, peptide, and nanoparticle-based approaches (1-3). However, until now only EphA2-targeting antibody-drug conjugates (ADC) have entered clinical development. For example, MEDI-547 is an EphA2-targeting ADC that displayed encouraging antitumor activity in preclinical models and progressed to phase I clinical testing in man. Here we describe the development of BT5528, a bicyclic peptide ("Bicycle") conjugated to the auristatin derivative maleimidocaproyl-monomethyl auristatin E to generate the Bicycle toxin conjugate BT5528. The report compares and contrasts the Pharmacokinetics (PK) characteristics of antibody and Bicyclebased targeting systems and discusses how the PK and payload characteristics of different delivery systems impact the efficacytoxicity trade off which is key to the development of successful cancer therapies. We show that BT5528 gives rise to rapid update into tumors and fast renal elimination followed by persistent toxin

Molecular imaging of cancers using probes specific for tumor-associated target proteins offers a powerful solution for providing information regarding selection of targeted therapy, patient stratification, and response to therapy. Here we demonstrate the power of bicyclic peptides as targeting probes, exemplified with the tumor-overexpressed matrix metalloproteinase MT1-MMP as a target. A bicyclic peptide with subnanomolar affinity towards MT1-MMP was identified, and its radioconjugate showed selective tumor uptake in an HT1080 xenograft mouse model. Proteolytic stabilization of the peptide by chemical modification significantly enhanced the in vivo tumor signal [from 2.5%ID/g to 12%ID/g at 1 hour post injection (p.i.)]. Studies using mouse xenograft models with different cell lines show a robust correlation between tumor signals and in vivo MT1-MMP expression levels. Fatty acid modification of the bicy-clic peptide extended its circulating half-life, resulting in increased tumor signals (36%ID/g at 6 hours p.i.). Comparative work with an equipotent radiolabeled MT1-MMP targeting antibody demonstrated starkly differential biodistribution and tumor accumulation properties, with the tumor signal slowly increasing to 6.2%ID/g within 48 hours. The rapid tumor penetration characteristics of bicyclic peptides, coupled with high potency and chemical versatility, thus offer high-contrast imaging probes for clinical diagnostics with compelling additional potential in targeted therapy.Significance: This work demonstrates the potential of bicyclic peptides as a platform for the development of highcontrast imaging probes for potential use in clinical cancer diagnostics and molecularly targeted therapeutics.

Plasma kallikrein, a member of the kallikrein-kinin system, catalyzes the release of the bioactive peptide bradykinin, which induces inflammation, vasodilation, vessel permeability, and pain. Preclinical evidence implicates the activity of plasma kallikrein in diabetic retinopathy, which is a leading cause of visual loss in patients suffering from diabetes mellitus. Employing a technology based on phage-display combined with chemical cyclization, we have identified highly selective bicyclic peptide inhibitors with nano- and picomolar potencies toward plasma kallikrein. Stability in biological matrices was either intrinsic to the peptide or engineered via the introduction of non-natural amino acids and nonpeptidic bonds. The peptides prevented bradykinin release in vitro, and in vivo efficacy was demonstrated in both a rat paw edema model and in rodent models of diabetes-induced retinal permeability. With a highly extended half-life of ∼40 h in rabbit eyes following intravitreal administration, the bicyclic peptides are promising novel agents for the treatment of diabetic retinopathy and diabetic macular edema.

We present a method for designing artificial receptors capable of binding with high affinity to a chosen target organic molecule. The primary sequence of the peptide is optimized to maximize its binding affinity. Our algorithm builds on a combination of molecular dynamics, semiflexible docking, and replica exchange Monte Carlo and performs simultaneous sampling in sequence and conformational spaces carefully selecting the degree of flexibility in the mutated peptides. The approach is used to design a decapeptide able to bind efavirenz. The calculated binding energy of the designed peptide (approximately -12 kcal/mol) was confirmed experimentally by fluorescence measurements. NMR spectroscopy confirmed the interactions between the peptide and the efavirenz molecule predicted by the algorithm.

Cathepsin K is a lysosomal cysteine protease that is highly and selectively expressed in osteoclasts, the cells which degrade bone during the continuous cycle of bone degradation and formation. Inhibition of cathepsin K represents a potential therapeutic approach for diseases characterized by excessive bone resorption such as osteoporosis. In order to elucidate the essential structural features for cathepsin K, a three-dimensional pharmacophore hypotheses were built on the basis of a set of known cathepsin K inhibitors selected from the literature using catalyst program. Several methods are used in validation of pharmacophore hypothesis were presented, and the fourth hypothesis (Hypo4) was considered to be the best pharmacophore hypothesis which has a correlation coefficient of 0.944 with training set and has high prediction of activity for a set of 30 test molecules with correlation of 0.909. The model (Hypo4) was then employed as 3D search query to screen the Maybridge database containing 59,000 compounds, to discover novel and highly potent ligands. For analyzing intermolecular interactions between protein and ligand, all the molecules were docked using Glide software. The result showed that the type and spatial location of chemical features encoded in the pharmacophore are in full agreement with the enzyme inhibitor interaction pattern identified from molecular docking.

Bicycles® are novel binding agents comprising small bicyclic peptides (1.5-3 KDa) constrained via a chemical scaffold, selected for high affinity and selectivity to targets of interest. MT1 (MMP14/MT1-MMP) is a membrane-associated metalloprotease overexpressed in many solid tumours and is implicated in tumor invasion and metastasis. MT1 expression positively correlates with poor prognosis. Phage libraries containing 1015 unique peptide sequences were post-translationally cyclized with thiol-reactive scaffold and used in an optimized, high-throughput selection process to identify Bicycles® to the hemopexin domain of MT1. Additional iterative rounds of directed phage based screening were used to optimize affinity and off-phage non-natural amino acids were introduced at select positions to improve plasma stability to generate the lead Bicycle binder. The lead anti-MT1 Bicycle was further modified with a sarcosyl spacer to form N241. N241 binds specifically to the hemopexin domain of MT1 with a Kd of approximately 2 nM with no binding observed to the catalytic domain of the protease nor to any of the related MMP family members tested. Importantly and in contrast to most antibodies, N241 binds with similar affinity to MT1 from multiple species including rodent, dog and non-human primate. Since the expected rapid tumor penetration and specific binding of these small peptidyl-binders makes them ideal for use in targeted delivery approaches, a series of Bicycle drug conjugates (BDCs) were prepared; N241 was conjugated to potent maytansinoid cytotoxics via linkers which varied in their cleavability. Though all the BDCs maintained high affinity for MT1, efficacy toward MT1-positive human tumor mouse xenografts varied with linker stability. BDCs with the most stable linkers were the least active suggesting that optimal tumor activation was obtained with linkers that could be cleaved more rapidly. Due to the rapid clearance and limited systemic exposure of these small-targeting BDCs, only the most labile linker showed toxicity in the mouse studies. Of the BDCs tested, BT1718, composed of N241 and DM1 conjugated via the SPP linker, demonstrated an optimal therapeutic index. Potent anti-tumor efficacy with BT1718 was observed across a panel of MT1-positive xenografts with complete tumor regressions observed in most models at doses that were well tolerated. In one example, HT-1080 fibrosarcoma subcutaneous xenografts were intraveneously treated with BT1718 when the tumor size had reached approximately 180 mm3. BT1718 given at 3 mg/kg once a week resulted in tumor stasis while BT1718 given at 10 mg/kg once a week or 3 mg/kg twice a week induced complete regression. In summary, BT1718, a highly active, targeted drug conjugate with unique pharmacological properties is a promising therapeutic candidate for the treatment of MT1-MMP-positive solid tumors. Citation Format: Helen Harrison, Gavin Bennett, Diane Blakeley, Amy Brown, Spencer Campbell, Liuhong Chen, Robert J. Lutz, Silvia Pavan, Katerine van Rietschoten, Daniel Teufel, Peter U. Park, Kevin Lee. BT1718, a novel bicyclic peptide-maytansinoid conjugate targeting MT1-MMP for the treatment of solid tumors: Design of bicyclic peptide and linker selection [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 5144. doi:10.1158/1538-7445.AM2017-5144