The Symmetric Tetravalent Sulfhydryl-Specific Linker NATBA Facilitates a Combinatorial "Tool Kit" Strategy for Phage Display-Based Selection of Functionalized Bicyclic Peptides.

The rigid conformation of constrained bicyclic peptides provides a number of advantages over larger protein-based ligands, including better chemical stability, enhanced tissue penetration, and a wider field of possible applications. Selective chemical modification strategies are able to extend the scope of applications not only in a therapeutic manner but also for the development of novel tools for protein capturing, bioimaging, and targeted drug delivery. Herein, we report the synthesis of an adamantane-based, symmetrical, tetravalent, sulfhydryl-specific peptide linker. We have developed an in vitro two-step modification strategy that allows the generation of differently functionalized bicyclic peptides. This "tool kit" strategy was applied to cyclize and functionalize a phage-encoded peptide library bearing the sequence CX6CX6C. After phage display against a model target, isolated peptides show strong consensus sequences, indicating target-specific binding. The newly developed symmetric tetravalent linker opens new avenues for the combinatorial selection and functionalization of bicyclic peptide ligands with affinity to virtually any target.

Introduction

The diverse biological activities and favorable properties of chemically constrained cyclic peptides make them an attractive format for applications in research and drug development.[1,2] Besides enhanced binding affinities and target specificities of constrained cyclic peptide scaffolds, they are able to interact with rather flat, featureless surfaces of proteins and therefore can efficiently target protein–protein interactions that are difficult to address by small molecules.[3] Furthermore, the oral bioavailability for the therapeutic application of small cyclic peptides has already been demonstrated.[4,5] In comparison to linear peptides, small cyclic peptides show higher proteolytic stability.[4] Heinis and Winter developed a phage display-based methodology for the selection of chemically modified peptides.[6−8] Inspired by the flexible regions of antibodies, their generated phage library presents peptides that contain two variable hexapeptide loops flanked by three cysteines.[8] Under physiological conditions, the comparatively high nucleophilicity of the sulfhydryl moiety of cysteine enables the selective reaction with certain electrophilic trivalent linker molecules like, e.g., tris(bromomethyl)benzene (TBMB) to generate bicyclic peptides.[9] To amplify the spectrum of potential applications, herein, we suggest to expand the one-step modification method on phage toward a two-step reaction by introducing the tetravalent linker N,N′,N″,N‴-(adamantane-1,3,5,7-tetrayl)tetrakis(2-bromoacetamide) (NATBA, 8). During the first step, the symmetrical tetravalent linker is efficiently modified by the three cysteine residues contained in every peptide of the phage library, leaving the fourth reactive position unmodified. In the second step of the protocol, this fourth position can be functionalized by a broad variety of sulfhydryl-bearing probes, warheads, linkers, tools, or drugs. Following this approach, the two-step modification allows the phage display of bicyclic peptide libraries that already bear a functional additive attached to the adamatane-based linker on phage during the selection process against any target. This tool kit strategy facilitates the development of novel peptide hybrids for therapeutic and diagnostic applications. The tetrahedral symmetry of the adamantane scaffold is an attractive feature that fulfills the requirements for the development of a tetravalent linker molecule. To retain the symmetry, the sulfhydryl-reactive moieties had to be attached to the four bridgehead carbon atoms of the adamantane structure. Because of its most suitable sulfhydryl-reactivity, the bromoacetyl group was chosen over other alternatives as the terminal reactive moiety forming a covalent thioether bond. To develop the two-step modification protocol, suitable test additives were required. Biotin and fluorescein as commonly used and easily accessible tools have been chosen as they provide various assay development possibilities. Commercially available 5(6)-carboxyfluorescein (Flc) and biotin are easy to equip with sulfhydryl groups by a simple one-step coupling reaction of 2-aminoethane-1-thiol to the respective carboxy moiety. These amide coupling reactions result N-(2-mercaptoethyl)-fluorescein-5(6)-carboxamide (FlcSH, 9) and N-(2-mercaptoethyl)biotinamide (BiotinSH, 10). Furthermore, we wanted to investigate the possibility to develop a two-step modification protocol that allows the attachment of larger much bulkier additives like peptides. For this purpose, the amphipathic cell-penetrating peptide FΦRRRR (FΦR4, Φ = l-2-napthylalanine)[10−12] was synthesized containing an N-terminal cysteine bound to 6-aminohexanoic acid (6-Ahx) as a linker that allows the flexible attachment to the peptide-linker construct. This additive is herein abbreviated as χFΦR4 (χ = Cys-6-Ahx).

Results and Discussion

Chemistry

The most reasonable synthetic route for the attachment of bromoacetyl moieties to the four bridgehead-carbon atoms of adamantane is the conjugation of respective bromoacetic acids via amide bond formation to 1,3,5,7-tetraaminoadamantane (7, see Scheme ). Synthetic routes toward functionalized adamantanes have been previously described.[13−15] The most common method for the effective fourfold activation of the bridgehead carbon atoms of adamantane is the aluminum chloride catalyzed bromination. Commercially available adamantane (1) was applied as starting material. The catalyzed bromination toward the corresponding 1,3,5,7-tetrabromoadamantane (2) was performed with a good yield of 68% using elemental bromine and aluminum chloride under reflux conditions for 10 h.[15] The following reaction for 3 was performed with sodium cyanate in dimethyl sulfoxide as solvent under direct UV (254 nm) light excitation by a 150 W low-pressure mercury vapor lamp. Different conditions were tried and the reaction was controlled by thin-layer chromatography (TLC) analysis. A reaction time of 6 h and stirring under argon atmosphere gave optimal results. Longer reaction times did not lead to significantly higher yields but to the formation of additional side products. Maintaining the reaction temperature below 25 °C resulted in significantly less formation of undesired side products. After purification, we obtained 3 in 89% yield. Nitrile hydrolysis for the corresponding carbonic acid (4) was performed under pressure and elevated temperatures using aqueous sodium hydroxide solution. After a reaction time of 6 h, the free acid 4 was precipitated by acidification of the reaction mixture. The hydrolysis led to a good yield of 70% for compound 4. The synthesis of 7 was performed via Curtius rearrangement. First, 5 was generated by refluxing 4 in thionyl chloride and a catalytic amount of dimethyl formamide. After 3 h of reaction, the solvent was removed to yield 5 without further purification and characterization. Compund 5 was converted into the azide 6 by the addition of sodium azide in cold acetone. After 2 h of stirring at room temperature, the excess sodium azide and generated sodium chloride were removed by filtration. The careful evaporation of the solvent yielded the tetraazide 6 as an oil. Due to the general instability of organic azides, 6 was not further characterized by 1H and 13C NMR analysis and directly applied in the following reaction. The corresponding 1,3,5,7-tetraisocynatoadamantane was generated by refluxing 6 in dichloroethane for 10 h. After removing the solvent, the tetrahydrochloride of 7 was generated by the addition of aqueous concentrated hydrochloric acid. After 1 h under reflux conditions, the aqueous phase was removed and the addition of aqueous sodium hydroxide resulted in crude 7. Sublimation of the solid at 200 °C and 5 mbar yielded pure 7 with a good yield of 71% with respect to the carbonic acid (4).

Scheme 1

Synthesis of Tetravalent Sulfhydryl-Specific Linker NATBA

(a) Br2, AlCl3, reflux, 10 h; (b) NaCN, dimethyl sulfoxide (DMSO), hν (254 nm), <25 °C, 6 h; (c) I: NaOH, H2O, 6 h; II: HCl; (d) SOCl2, dimethylformamide (DMF), reflux, 3 h; (e) NaN3, acetone, RT, 2 h; (f) I: C2H4Cl2, reflux, 10 h; II: HClconc, 1 h; reflux III: NaOH, H2O, RT, 5 min; (g) bromoacetic acid, PyOxim, DIPEA, DMF, RT, 30 min.

First attempts to synthesize the desired final linker molecule 8 by the reaction of 7 with 2-bromoacetyl chloride, 2-bromoacetyl bromide, 2-chloroacetyl chloride, or 2-chloroacetyl bromide in different solvents and under varying conditions failed. One hypothesis for this is the possible formation of highly reactive ketenes out of acyl halides bearing at least one hydrogen at the α-carbon atom in the presence of amines.[16] To overcome this problem, the desired amide formation between the bromoacetic acids and 7 was evaluated using different coupling reagents, like HBTU, PyBOP, PyBroP, PyOxim, bases, and solvents. Among all tested conditions, the reaction with PyOxim in dimethyl formamide as solvent and diisopropylamine as a base gave the best yields for 8 with 39%. First, the 4.1-fold amount of the respective haloacetic acid was activated for 20 min with an equimolar amount of PyOxim in the presence of an excess diisopropylamine as a base. Subsequently, the activated acid mixture was added to a suspension of 7 in dimethyl formamide. After a reaction time of 30 min, no further starting material could be detected by TLC analysis. Longer reaction times did not lead to significantly higher yields but to the formation of additional side products. The use of more equivalents or a stepwise addition of activated acid did not change the results. Subsequently, the reaction mixture was cooled to 0 °C and directly purified over a C8-modified reversed-phase silica column by using a water/acetonitrile gradient. Product-containing fractions were directly flash frozen and lyophilized yielding the desired compound 8 with an overall yield of approximately 12% with respect to adamantane (1).

Development of the Two-Step Modification Strategy

The implementation of NATBA as a linker in phage display-based screening of bicyclic peptides requires a selective and almost quantitative chemical modification of the peptides presented by phages under physiological conditions.[6,8]

It is crucial for the success of the phage display to retain phage infectivity after peptide modification and the variations of reaction conditions, like the choice of solvent, pH, and temperature, are limited by phage stability. For the investigation of the sulfhydryl reactivity of the newly synthesized linker NATBA, we chose a matrix-assisted laser desorption ionization time-of-flight (MALDI-TOF) controlled strategy for the optimization of the reaction conditions. As a positive control, we monitored the reaction process of the cyclization linker TBMB with the test peptide TPA (NH2-ACEGMINSCEKSDYECG-CONH2, 1839.1 Da) under reaction conditions published by Heinis et al. (20 mM NH4CO3, 5 mM EDTA, pH 8, 30 °C) over time by MALDI-TOF mass spectrometry.[9] In less than 5 min, the linear test peptide TPA was almost quantitatively (>95%) cyclized to form the desired bicycle peptide TPA-TBMB. However, the TPA-TBMB bicyclic peptide continued to react with excess TBMB (1.5 equiv) upon extended reaction time (>10 min) (Figure S2, Supporting Information). The reaction of NATBA was investigated under varying reaction conditions. Different buffers, pH values (6–8), acetonitrile concentrations, and temperatures were tested. The best results were achieved using a HEPES buffer at pH 8, 15% acetonitrile, and 30 °C (see Figure ). At pH 6, the reaction proceeded too slowly. After 1 h of reaction, about 50% conversion could be detected. As expected, the reaction rate of NATBA increased when choosing the pH values closer to the pKS of cysteine. At pH 7, the reaction was nearly completed after 30 min. This optimization process led to pH 8 for buffering and a reaction time of less than 10 min.

Figure 1

Reaction of NATBA with test peptide TPA (NH2-ACEGMINSCEKSDYECG-CONH2, 1839.1 Da) measured by MALDI-TOF over a time of 60 min at different pH. Reaction conditions: (A) 20 mM HEPES, 15% acetonitrile, pH 8.0, 30 °C, 0.5 mM TPA, and 1.5 equiv NATBA. (B) 20 mM HEPES, 15% acetonitrile, pH 7.0, 30 °C, 0.5 mM TPA, and 1.5 equiv NATBA. (C) 20 mM HEPES, 15% acetonitrile, pH 6.0, 30 °C, 0.5 mM TPA, and 1.5 equiv NATBA. Molecular weight TPA-NATBA: 2276.4 Da. Annotation: Cross-linked, dimerized/trimerized, etc., peptides using this protocol were not observed.

Thus, for the second reaction step, the respective additives (FlcSH, BiotinSH, cFΦR4, see Figure ) were added after step 1 (reaction of test peptide TPA with NATBA for 10 min at pH 8 and 30 °C using 20 mM HEPES and 15% acetonitrile). Dependent on the added molecule, the reaction yielding the desired functionalized bicycle took between 30 and 60 min. The best overall conversions were achieved by using 1.05 equiv of NATBA and 1.0 equiv of respective additive minimizing the formation of multiple conjugation products formed by excess of additives or linker molecules. At the beginning of the development phase, the approach to precouple NATBA (8) with a functionalization molecule was also intended. However, the synthesis turned out to be particularly difficult because the multicoupled byproducts are very difficult to separate from the singly coupled product. Due to very low yields, this approach was not pursued. It should be noted that the reaction of a planar, trivalent linker such as TBMB with a peptide with two variable cysteine-free sequences flanked by three cysteines can only generate one single species of bicyclic peptides. In comparison to that, the reaction between a tetravalent linker and a peptide of the same type could result in two stereo isomers, most likely giving a mixture of these two species. Starting from a peptide library of the format CX6CX6C, where a fourth cysteine can appear in any of the randomized amino acid positions (X), Chen et al.[17] generated a combinatorial library of bicyclic peptides by oxidization of these four cysteines to two pairs of cystines. During the selection process, a strong enrichment of peptides with four cysteines was observed, yielding binders to protein targets in the low micromolar range. This approach demonstrated that also phage display of bicyclic peptides with two or even more different possible regioisomers per phage (about five peptide copies are displayed per phage[9]) is able to generate potent peptide binders.[17]

Figure 2

Two-step reaction of NATBA with test peptide TPA (NH2-ACEGMINSCEKSDYECG-CONH2, 1839.1 Da) and the additives χFΦR4, FlcSH, and BiotinSH measured by MALDI-TOF. Reaction conditions: 20 mM HEPES, 15% acetonitrile, pH 8.0, 30 °C, 0.5 mM TPA, and 1.05 equiv NATBA. TPA-NATBA: 2276.4 Da. TPA-NATBA-FlcSH: 2629.9 Da, 2654.0 Da [M + Na+]+, TPA-NATBA–BiotinSH: 2497.5 Da, and TPA-NATBA-χFΦR4: 3396.9 Da.

Phage Affinity Selection of Functionalized Peptides

The phage display-based methodology for the selection of chemically modified peptides developed by Heinis and Winter[6,8] served as a framework for the implementation of the newly developed two-step tool kit concept. The phage library with a complexity of 8.8 × 107 different clones presenting linear peptides of the format ACX6CX6CG-phage (X = all natural occurring amino acids except cysteine) was prepared with minor modifications as described in the literature (see Supporting Information).[9] In general, it has to be assumed that after the preparation of the phages, the cysteine-rich peptide library is partly oxidized. Therefore, it is essential to reduce the potentially oxidized peptides before modification. Heinis and co-workers have implemented this reduction of phages via tris(2-carboxyethyl)phosphine (TCEP) into their protocol.[7−9] This includes a stepwise dilution procedure of the reducing agent by using centrifugal filters. We developed a faster and more effective TCEP removal step by purifying the reduced phages over a Sephadex desalting column. Subsequently, the linear peptides were selectively modified by the two-step modification under optimized conditions using NATBA as a linker and FlcSH as an additive. The phage-encoded bicyclic peptide library was subjected to three iterative rounds of affinity selection against the model target (rabbit anti-goat IgG (H + L) Superclonal Secondary Antibody) using this modification protocol. After each selection, the captured phages were eluted by treatment with low pH buffer and rescued by infecting E. coli. After the third iterative round, individual clones were selected and sequenced revealing the respective amino acid sequences (see Figure A). We were pleased to find strong consensus motifs within the resulting peptides sequences as a clear indication for target-specific binding. Most of the similarities are found in the N-terminal part of the first loop, consisting predominately of Leu, His, and Phe/Tyr. In the second loop, the highest conserved position is likewise the most N-terminal amino acid next to the central Cys, showing a strong preference for polar or acidic residues (Ser, Asp, Gln). None of the isolated sequences were found with an abundance higher than 3. This may indicate that the affinities are in a similar range. To confirm the affinity against the model target, a magnetic bead-based pulldown experiment was performed. Six individual selected sequences representing different sequence motifs were synthesized and functionalized with BiotinSH (10). As a negative control, a sequence (JK38) sharing no observable similarity to any of the chosen motifs was used. The BiotinSH-functionalized peptides were captured by streptavidin immobilized on magnetic beads (Figure B). After incubation with the target protein and five washing cycles, each pulldown experiment was analyzed by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) (Figure C). For all of the analyzed sequences, except the negative control, the light and the heavy chain of the rabbit anti-Goat IgG antibody was clearly detected as an indicator of the successful capturing by the respective functionalized peptide.

Figure 3

(A) Peptides isolated in phage selections against recombinant rabbit anti-goat IgG. Amino acid sequences are displayed using CLC-sequence viewer[18] and colored in the Rasmol color scheme.[19] The frequency with which each peptide sequence was found is noted under “abundance”. (B) Graphical representation of the pulldown strategy. (C) SDS-PAGE of the pulldown experiment performed with the sequences as indicated. Sequence JK38 was applied in the pulldown experiment as a negative control. The molecular weight (MW) of the IgG heavy chain is approx. 50 kDa. The MW of the light chain is approx. 25 kDa. The band at 17 kDa represents the ubiquitously high levels of streptavidin (16.8 kDa) washed from the magnetic beads after the pulldown experiment.

Conclusions

We were able to demonstrate that bicyclic peptides can efficiently be formed and functionalized by different additives in two consecutive reaction steps using the herein presented tetravalent linker NATBA (8) and applying the optimized reaction conditions provided in our protocol. Implementing this novel NATBA modification concept into phage display, we could finally isolate functionalized bicyclic peptides that bind recombinant rabbit anti-goat IgG antibody as a model target. The successful capturing of the target protein in pulldown experiments by six different selected peptides containing biotin suggests furthermore that the spatial orientation of the functionalized peptide construct might allow exchanges at the fourth position of the linker without a complete loss of affinity. However, current data do not allow a clear statement how far functionalization affects final affinity. The synthesis of the sulfhydryl-reactive NATBA (8) described here provides a general route toward the symmetric tetravalent linkers that enables, in addition to the presented concept, also the generation of globular tricyclic peptides. Further use of this approach is currently under investigation. In summary, NATBA is the center of a tool kit directly linking affinity selection by phage display to functionalization of the bicycles within the selection process.

Experimental Section

Chemicals and Reagents

All the chemicals and reagents were obtained from commercial sources and used as received unless otherwise noted.

1,3,5,7-Tetrabromoadamantane (2)

This reaction was performed with minor modifications as described in the literature.[20] A mixture of anhydrous aluminum chloride (30 g, 0.23 mol) and bromine (15 mL, 0.39 mol) was cooled to 0 °C. Adamantane (1) (7.5 g, 55.1 mmol) was added carefully in small portions through a solid addition funnel. Subsequently, the reaction mixture was slowly heated up and refluxed. HBr gas that evolved was passed through a second flask containing concentrated aqueous sodium bisulfite. When no more HBr gas evolved (about 10 h), small portions of cooled saturated sodium bisulfite solution (100 mL) and aqueous HCl (4 N, 100 mL) were added to the reaction mixture under intense cooling. The resulting dark yellow oily precipitate was taken up in acetone (100 mL). To the resulting dark brown solution, water was added in steps of 150 mL each to precipitate the brominated adamantane. The resulting crude product was purified by column chromatography on silica (eluent: n-hexane) yielding the desired 1,3,5,7-tetrabromoadamantane (2) as a light-brownish solid. Recrystallization from acetone/water yielded the product with a purity of 98%. Yield: 16.9 g, 68% (lit.: 61%[20]). 1H NMR (400 MHz, CDCl3) δ: 2.64 (s, 12H) ppm. 13C NMR (100 MHz, CDCl3) δ: 54.8, 54.6 ppm. UVmax = 262 nm.

1,3,5,7-Tetracyanoadamantane (3)

1,3,5,7-Tetracyanoadamantane (3) was prepared with minor modifications according procedures described in the literature.[20] 1,3,5,7-Tetrabromoadamantane (2) (2.14 g, 4.74 mmol) and sodium cyanide (3.69 g, 75.29 mmol) were dissolved in 150 mL degasified dimethyl sulfoxide in a cylindrical quartz tube. The mixture was irradiated under argon at 254 nm for 6 h (150 W low-pressure mercury vapor lamp). After irradiation, the solvent was removed by distillation at reduced pressure. The brownish residue was washed with water (500 mL) and the solid precipitate was filtered off. The filtrate was directly collected in a solution of 6% sodium hypochlorite. The crude product was purified by column chromatography on silica. Unreacted starting material and partially cyanated products were removed by dichloromethane. A mixture of dichloromethane/acetone (1:4) served to elute the desired product. Recrystallization from acetone/water yielded the desired 1,3,5,7-tetracyanoadamantane (3) as a light-brownish solid. Yield: 990 mg, 89% (lit.: 73%[20]). 1H NMR (400 MHz, DMSO-d6) δ: 2.39 (s, 12H) ppm. 13C NMR (100 MHz, DMSO-d6) δ: 54.8, 54.6, 120.9 ppm. Mp: >400 °C. HRMS-FAB: calc.: 236.27 [M + H+]; found: 236.32; UVmax = 257 nm.

1,3,5,7-Tetracarboxyadamantane (4)

1,3,5,7-Tetracyanoadamantane (3, 990 mg, 4.19 mmol) was suspended in an aqueous sodium hydroxide solution (15 wt %, 30 mL). The reaction mixture was heated under stirring in a pressure tube at 135 °C for 6 h. The initially suspended starting material dissolved within 3 h. After cooling to room temperature, the slight brownish solution is acidified with concentrated HCl. The precipitation of the product started immediately. After filtration under reduced pressure, the colorless solid was redissolved in water (20 mL) under dropwise addition of sodium hydroxide solution (0.1 M). After acidifying the solution using concentrated HCl to pH = 1, the product was allowed to recrystallize at 4 °C for 1 h. Subsequently, the light-brownish solid was filtered under reduced pressure and dried at 100 °C for 10 h yielding 1,3,5,7-adamantanetetracarboxylic acid (4) as light-brownish solid. Yield: 915 mg, 70%. 1H NMR (400 MHz, DMSO-d6) δ: 1.79 (s, 12H), 12.43 (s, 4H) ppm. 13C NMR (100 MHz, DMSO-d6) δ: 38.5, 40.8, 176.7 ppm.

1,3,5,7-Tetraaminoadamantane (7)

A mixture of 1,3,5,7-tetracarboxyadamantane (4) (495 mg, 1.58 mmol), thionyl chloride (5 mL), and a catalytic amount of dimethyl formamide (1 drop) was refluxed for 3 h. Within the first 1.5 h, the initially suspended starting material dissolved. The thionyl chloride was removed under heat (65 °C), stirring, and argon stream. Remaining dimethyl formamide was removed under high vacuum, yielding adamantane-1,3,5,7-tetracarbonyl tetrachloride (5) as a colorless solid in quantitative yield (609 mg, 98%). To the remaining solid an ice-cold suspension of sodium azide (616 mg, 9.48 mmol) in acetone (5 mL) was added under stirring and cooling to 0 °C. After 10 min, the cooling bath was removed and the reaction mixture was allowed to warm up to room temperature. After 2 h of stirring, the excess of insoluble sodium azide and sodium chloride were filtered and the filtrate evaporated under reduced pressure to yield adamantine-1,3,5,7-tetracarbonyl tetrazide (6) as a red-brownish oil (529 mg, 81%). To the oil, 1,2-dichloroethane (15 mL) was added and the reaction mixture was refluxed under stirring for 10 h. Subsequently, the solvent was removed under reduced pressure and concentrated hydrochloric acid (20 mL) was added to the light-brownish crude 1,3,5,7-tetraisocyanatoadamantane and refluxed for 1 h. Then, the acidic aqueous phase was removed under reduced pressure to yield 1,3,5,7-tetraaminoadamantane tetrahydrochloride (433 mg, 80%). To the slightly brownish residue dissolved in water (5 mL), sodium hydroxide (284 mg, 7.1 mmol) was added. Subsequently, the solvent was removed under reduced pressure. Sublimation of the brownish crude solid product at 250 °C and 5 mbar yielded pure 1,3,5,7-tetraaminoadamantane (7) as a colorless solid. Yield: 221 mg, 71%, colorless solid. 1H NMR (400 MHz, DMSO-d6) δ: 1.12 (s, 12H) ppm. 13C NMR (100 MHz, DMSO-d6) δ: 50.7, 52.9 ppm.

N,N′,N″,N‴-(Adamantane-1,3,5,7-tetrayl)tetrakis(2-bromoacetamide) (8)

To a suspension of 1,3,5,7-tetraaminoadamantane (7) (50.6 mg, 0.26 mmol) in dry DMF (1 mL) was added activated bromoacetic acid solution (2.5 mL) under stirring at room temperature directly followed by dry N,N-diisopropylethylamine (221 mL, 1.30 mmol). The activated bromoacetic acid solution was prepared as follows: to a solution of PyOxim (686.5 mg, 1.30 mmol) in dry DMF (2 mL) was added bromoacetic acid (180.7 mg, 1.30 mmol) under stirring at room temperature and stirred for 20 min. The suspended 1,3,5,7-tetraaminoadamantane (7) dissolves within 1 min after the addition of the activated bromoacetic acid and N,N-diisopropylethylamine. After 0.5 h of stirring, the crude reaction mixture was directly purified in 1 mL steps over a C8-modified reversed-phase silica column (PP-30C8/35G, 30 mm PP, 300, Interchim, France) by using the following acetonitrile (ACN)/ddH20 gradient method: 0 min −8% ACN, 20 min −25% ACN. Product peak is to be expected at approx. 20% ACN. The product-containing fractions were directly flash frozen and subsequently lyophilized. Yield: 68.5 mg, 39%. 1H NMR (400 MHz, DMSO-d6) δ: 2.12 (s, CH2 12H), 3.79 (s, CH2Br, 8H); 8.20 (s, NH, 4H) ppm. 13C NMR (100 MHz, DMSO-d6) δ: 30.8, 42.6, 52.9, 166.0 ppm. LC-MS (ESI): 702.09 [M + Na]+, 679.9 [M – H]−.

N-(2-Mercaptoethyl)-fluorescein-5(6)-carboxamide (9)

To a mixture of 5/6-carboxyfluorescein (100 mg, 2.66 mmol), 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide hydrochloride (760 mg, 4 mmol), and N-hydroxysuccinimide (460 mg, 4 mmol) in dry dimethylformamide (7 mL) was added cystamine (610 mg, 0.4 mmol). After stirring for 24 h at room temperature, the crude product was precipitated by the addition of water (20 mL). After filtration, the solid was washed twice with cold water (25 mL) and resuspended in a mixture of methanol and water (8:2, 20 mL). Under stirring was added tris(2-carboxyethyl)phosphine (154.23 mg, 0.530 mmol). After 8 h of reaction at room temperature, the solvent was removed under reduced pressure and the crude product was purified by silica column chromatography (eluent:dichloromethane/methanol). Yield: 920 mg (80%). 1H NMR (200 MHz, DMSO-d6) δ: 1.35 (s, SH, 1H), 2.63 (t, HSCH2, 2H), 3.20–3.40 (m, HNCH2 2H), 6.60–6.75 (m, Harom. (2,4,5,7), 4H), 6.83 (s, Harom. (1,8), 2H), 7.52 (d, J = 6 Hz, Harom. 70, 1H), 7.81–8.40 (m, Harom. 4′, 4″, 5′, 6′, 6″, 7″, 6H), 8.6 (s, NH, 1H), 10.31 (s, OH, 2H) ppm. 13C NMR (50 MHz, DMSO-d6) δ: 25.0, 43.3, 85.0, 97.9, 102.6, 109.44, 123.0, 123.7, 124.7, 126.8, 128.7, 129.5, 129.8, 135.1, 136.4, 140.9, 152.2, 155.1, 158.2, 160.0, 165.6, 168.6 ppm; LC–MS: 436.1 [M + H]+, 434.1 [M – H]−.

N-(2-Mercaptoethyl)biotinamide (10)

To a suspension of biotin (50.0 mg, 0.2 mmol), dicyclohexylcarbodiimid (46.45 mg, 0.23 mmol), and N-hydroxysuccinimide (25.91 mg, 0.23 mmol) in dimethylformamide (2 mL) was added cystamine (15.58 mg, 0.1 mmol). After 16 h, the precipitate was filtered off and washed with DCM (20 mL). Subsequently, the solid was dried and redissolved in a mixture of methanol and water (8:2, 20 mL). Under stirring (tris(2-carboxyethyl)phosphine) hydrochloride (86 mg, 0.3 mmol, 286.65 g/mol) was added. After 2 h of stirring, the solvent was removed under reduced pressure and the crude product was purified by preparative HPLC (C18, 5 μm, 20 × 250 mm, Reprosil100, Dr. Maisch, Ammerbuch, Germany) (eluent: ddH2O + 0.1% TFA and 80% ACN/ddH2O + 0.1% TFA). Product-containing fractions were directly flash frozen and lyophilized, yielding the desired N-(2-mercaptoethyl)biotinamide as a colorless solid. Yield: 111.7 mg (90%). 1H NMR (300 MHz, DMSO-d6) δ: 1.25–1.68 (m, CH2CH2CH2, 6H), 2.07 (t, J = 6 Hz, CH2C(O)NH, 2H), 2.32 (t, J = 9 Hz, CHCH2, 1H), 2.47–2.60 (m, CH2SH, 2H), 2.83 (t, J = 9 Hz, CHCH2bS, 1H), 3.07–3.11 (m, CHCHS, 1H), 3.11–3.20 (m, C(O)NHCH2, 2H), 4.13 (dd, J = 3 Hz, NHCHCH, 1H), 4.31 (dd, J = 3 Hz, NHCHCH2, 1H), 6.42 (s, C(O)(NH)2, 2H), 7.95 (t, J = 6 Hz, C(O)NH, 1H) ppm. 13C NMR (75 MHz, DMSO-d6) δ: 24.0, 25.7, 28.5, 28.7, 35.6, 40.3, 42.5, 55.9, 59.7, 61.5, 163.2, 172.6 ppm. MS (ESI): 302.9 [M + H]+, 326.0 [M + Na]+.

Chemical Synthesis of Peptides

Peptides with free N-terminus and amidated C-terminus were synthesized on a 50 μmol scale by standard Fmoc solid-phase chemistry on an automated peptide synthesizer (Syro I, MuliSynTech GmbH, Witten, Germany). After deprotection and cleavage from the 4-(2′,4′-dimethoxyphenyl-Fmoc-aminomethyl)phenoxymethyl (Rink amide) resin (TentaGel HL RAM, RAPP Polymere, Tuebingen, Germany) with trifluoroacetic acid (85 vol %, 5.5 wt % phenol, 4.5 vol % triisopropylsilane, 5 vol % H2O), the resulting deprotected peptides were precipitated three times in cold diethylether (−20 °C). Subsequently, the crude peptides were purified by reversed-phase HPLC (Interchim, PuriFlash 4250) using a Reprosil100 column (C18, 5 μm, 20 × 250 mm2, Dr. Maisch, Ammerbuch, Germany) and a linear gradient elution with a mobile phase composed of eluent A (99.9 vol % H2O and 0.1 vol % trifluoroacetic acid) and eluent B (80 vol % acetronitrile, 20 vol % H2O, and 0.1 vol % trifluoroacetic acid) at a flow rate of 15 mL/min. The molecular masses were determined by MALDI-TOF mass spectrometry (Bruker Daltonics—autoflex II, MA). Cyclization with NATBA: The peptide (1 mM) dissolved in 3 mL 70% buffer R (20 mM HEPES, pH 8)/30% acetonitrile was reacted with 1.1 equiv NATBA dissolved in 1 mL acetonitrile under shaking at 30 °C. After 10 min, of reaction, 2 equiv of 10 dissolved in 1 mL of 50% buffer R (20 mM HEPES, 5 mM EDTA, pH 8)/50% acetonitrile were added and the reaction was incubated under gentle shaking for further 60 min at 30 °C. Chemically modified peptides were purified as described above. The purity of the modified and unmodified peptides were assessed by analytical RP-HPLC (Agilent 1100 system, Agilent Technologies, Santa Clara, CA), using a C18 column and the same buffer system as for preparative RP-HPLC.

Peptide Cyclization and Mass Spectrometric Analysis

Test peptides were dissolved in the respective buffer system and placed in a thermoshaker (Bioer Technology) at 30 °C. After 10 min of incubation, the respective amount (1.5 equiv) of linker dissolved in a volume of 100 μL acetonitrile was added to 500 μL test peptide solution (0.5 mM) under shaking. The progress of the reaction was determined by MALDI-TOF mass spectrometry at different times (1, 5, 10, 30, and 60 min). Taken samples were directly mixed with the same volume of matrix solution (20 mg α-cyano-4-hydroxycinnamic acid (α-CHCA) in 1 mL 50:50 H2O/ACN with 0.1% TFA trifloroacetic acid (final conc.)), spotted on target plate and measured by MALDI-TOF mass spectrometry (Bruker Daltonics—autoflex II, MA).

Preparation of the 6 × 6 Phage Library

The 6 × 6 phage library that presents peptides of the format ACX6CX6G was generated with modifications as described in the literature (see Supporting Information).[9]

Phage Selection of Functionalized Modified Peptides

Phage display was performed with modifications as described in the literature.[9]E. coli TG1 glycerol stock cells containing the phage library were used to inoculate 500 mL of 2xYT media containing 30 mg/mL until an OD600 of 0.1 was reached. After incubation for 16 h at 30 °C, the E. coli TG1 cells were removed from the phage-containing supernatant by centrifugation for 30 min at 6000 rpm and 4 °C. The phages were precipitated by the addition of 0.2 volume of precipitation buffer (PEG buffer, 20% PEG-6000, 2.5 M NaCl). After the addition of the PEG buffer, the phage suspension was cooled on ice for 60 min and subsequently centrifuged for 30 min at 7000 rpm and 4 °C. After careful removal of the supernatant, the resulting phage pellets were resuspended in 10 mL ice-cold buffer R (20 mM HEPES, 5 mM EDTA, pH 8) and again centrifuged for 30 min at 4000 rpm and 4 °C. The phage-containing supernatant was carefully transferred into a new 50 mL falcon tube. To the clear phage solution was added TCEP with a final concentration of 1 mM and the reduction was incubated at 42 °C for 1 h. To remove TCEP, the reduced phage solution was purified twice over a desalting column (HiPrep 26/10 Desalting, GE Healthcare). The resulting phage-containing fraction was adjusted to 32 mL with buffer R and 4 mL of 100 mM NATBA solution in acetonitrile was added to obtain a final linker concentration of 11.11 μM/L and incubated at 30 °C for 10 min under gentle shaking followed by the direct addition of 4 mL 500 μM FlcSH (9) in a mixture of 50% acetonitrile/H2O and further incubation at 30 °C for 1 h. The chemically modified phages were subsequently precipitated by the addition of 0.2 volume of buffer PEG, cooled down on ice for 30 min, and centrifuged at 4700 rpm for 30 min at 4 °C. The phage pellet was dissolved in 3 mL buffer W1 (10 mM Tris–HCl, 150 mM NaCl, 10 mM MgCl2, 1 mM CaCl2, pH 7.4). Biotinylated Rabbit anti-Goat IgG (H + L) Superclonal Secondary Antibody (ThermoFischer) was immobilized on magnetic streptavidin beads (Dynabeads M-280 Streptavidin, Invitrogen) following the protocol provided in the Supporting Information. Negative control without the antigen was prepared by addition of buffer W1 instead of antigen. Subsequently, the magnetic beads were washed three times with 0.5 mL buffer W1 and incubated for 30 min at room temperature in 300 mL buffer W1 supplemented with 150 mL buffer W2 (buffer W1 + 3% BSA, 0.3% Tween 50). During that step, the chemically modified phages, stored in 3 mL buffer W1, were blocked at room temperature by the addition of 1.5 mL buffer W2 for 30 min. The blocked bead suspension (0.45 mL) and 2.25 mL of the blocked phage suspension were mixed together and incubated for 30 min on a rotating wheel at room temperature. Same was performed for the negative control. The beads were washed eight times with buffer W3 (buffer W1 + 0.1% Tween 50) and twice with buffer W1. The phages were eluted by incubation with 100 μL of buffer E (50 mM gylcine, pH 2) for exactly 5 min and then directly transferred into 50 μL of buffer N (1 M Tris, pH 8) for neutralization. The eluted phages were added to 25 mL of E. coli TG1 cells at OD600 of 0.4 for 90 min at 37 °C. After centrifugation for 5 min at 4000 rpm and 4 °C, the cell pellets of positive and negative experiments were plated on each two large 2xTY/chloramphenicol (30 mg/mL) plates. For each round, the input and output phage titers were determined. Second and third rounds of panning were performed following the same procedure but using in the second round instead of streptavidin beads neutravidin-coated magnetic beads. Magnetic neutravidin beads were prepared by reacting 1 mg neutravidin (Pierce) with 0.5 mL tosyl-activated magnetic beads (Dynabeads M-280 Tosyl-activated, Invitrogen) according to the supplier’s instructions. After the second and third round of selections, individual clones were picked and amplified. The extracted plasmid DNA of each clone was sequenced using primer seqba (Eurofins, Germany).

Pulldown Assay

Individual selected peptide sequences were synthesized and functionalized with NATBA as a linker and BiotinSH (10) as an additive. The required amount of magnetic streptavidine beads (250 mg per sample) was washed three times with 1 mL phosphate-buffered saline (PBS) and spitted into the respective number of experiments. After adding to each sample 1 mg of the respective NATBA–BiotinSH (10) functionalized peptide dissolved in a minimum amount of PBS, the bead suspension was incubated for 15 min at room temperature on a wheel shaker. Subsequently, the suspension was washed five times with each 1 mL PBS and transferred into new 1.5 mL tubes. To each sample, 2 μL of rabbit anti-goat IgG (H + L) Superclonal Secondary Antibody (1 mg/mL, ThermoFischer) was added. After incubation for 1 h at room temperature on a wheel shaker, the bead suspension was washed five times with each 1 mL PBS and transferred into new 1.5 mL tubes. Washed beads were analyzed by SDS-PAGE.