(TRF2) Protein–Protein Interaction in the Shelterin ... - ACS Publications

Article pubs.acs.org/jmc

Design of High-Affinity Stapled Peptides To Target the Repressor Activator Protein 1 (RAP1)/Telomeric Repeat-Binding Factor 2 (TRF2) Protein−Protein Interaction in the Shelterin Complex Xu Ran,†,∥ Liu Liu,‡,∥ Chao-Yie Yang,‡,∥ Jianfeng Lu,‡,∥ Yong Chen,⊥,# Ming Lei,⊥,# and Shaomeng Wang*,†,‡,§,∥ Department of Medicinal Chemistry, ‡Department of Internal Medicine, §Department of Pharmacology, ∥Comprehensive Cancer Center, ⊥Department of Biological Chemistry, and #Howard Hughes Medical Institute, University of Michigan, Ann Arbor, Michigan 48109, United States

Downloaded via TULANE UNIV on January 21, 2019 at 15:45:59 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

†

S Supporting Information *

ABSTRACT: Shelterin, a six-protein complex, plays a fundamental role in protecting both the length and the stability of telomeres. Repressor activator protein 1 (RAP1) and telomeric repeat-binding factor 2 (TRF2) are two subunits in shelterin that interact with each other. Small-molecule inhibitors that block the RAP1/TRF2 protein−protein interaction can disrupt the structure of shelterin and may be employed as pharmacological tools to investigate the biology of shelterin. On the basis of the cocrystal structure of RAP1/TRF2 complex, we have developed first-in-class triazole-stapled peptides that block the protein−protein interaction between RAP1 and TRF2. Our most potent stapled peptide binds to RAP1 protein with a Ki value of 7 nM and is >100 times more potent than the corresponding wild-type TRF2 peptide. On the basis of our high-affinity peptides, we have developed and optimized a competitive, fluorescence polarization (FP) assay for accurate and rapid determination of the binding affinities of our designed compounds and this assay may also assist in the discovery of non-peptide, small-molecule inhibitors capable of blocking the RAP1/TRF2 protein−protein interaction.

■

INTRODUCTION

Shelterin is a six-protein complex that protects telomeres from DNA repair mechanisms and regulates telomerase activity.1−9 In mammalian cells, the shelterin complex consists of TRF1, TRF2, TIN2, RAP1, TPP1, and POT1 proteins.10 TRF1, TRF2, and POT1 recognize TTAGGG DNA repeats of the telomere and also interconnect with TIN2, TPP1, and RAP1 proteins to form the shelterin complex. The shelterin complex plays a fundamental role of protecting human telomeres, and components of the shelterin complex, including a number of protein−protein interactions, are potential new therapeutic targets for the treatment of human cancers and other diseases. Within the shelterin complex, TRF2 and TRF1 form a heterodimer that binds to DNA. Additionally, TRF2 enjoys protein−protein interactions with TIN1 and RAP1 proteins through its two different domains. The cocrystal structure of the RAP1/TRF2 complex (Figure 1, PDB code 3K6G)11 provides atomic details of their interactions. The cocrystal structure reveals that the RAP1 binding motif (RBM) of TRF2, which consists of two three-turn short helices, named α1 and α2, and four unstructured residues linked to these two helices, interacts with a six-helical bundled RAP1 C-terminal (RCT) domain. The RAP1/TRF2 complex structure further shows that the α1 helical segment of TRF2 is a major contributor to the RAP1/TRF2 interaction, by binding deeply inside the pocket of RAP1 and enjoying extensive hydrophobic contacts with RAP1. © 2015 American Chemical Society

Figure 1. Crystal structure of RAP1/TRF2 complex (PDB code 3K6G). TRF2 RBM (yellow) employs two α helices to interact with RAP1 RCT (green).

Our computational analysis of the interaction between the α1 helical segment of TRF2 and RAP1 suggested the intriguing possibility of design and development of high-affinity peptidebased compounds or even non-peptide small-molecule inhibitors capable of blocking the RAP1/TRF2 protein− Received: September 19, 2015 Published: December 16, 2015 328

DOI: 10.1021/acs.jmedchem.5b01465 J. Med. Chem. 2016, 59, 328−334

Journal of Medicinal Chemistry

Article

Table 1. Binding Affinities of Truncated TRF2 Peptides to RAP1 Protein Determined in Our Optimized FP Competitive Binding Assay 1 2 3 4 5 6

sequence

IC50 (μM)

Ki (μM)

Ac-281TTIGMMTLKAAFKTLS296-NH2 Ac-281TTIGMKTLKAAFKTLS296-NH2 Ac-281TTIGMKTLKAAFKTL295-NH2 Ac-281TTIGMKTLKAAFKT294-NH2 Ac-282TIGMKTLKAAFKTL295-NH2 Ac-283IGMKTLKAAFKTL295-NH2

9.3 ± 0.4 8.3 ± 0.4 27.0 ± 0.1 >200 90.3 ± 4.2 182 ± 2

2.0 ± 0.1 1.8 ± 0.1 5.8 ± 0.1

To enhance the solubility of peptide 1, we mutated its solvent exposed residue Met286 to lysine, yielding peptide 2 (Table 1). Peptide 2 binds to RAP1 with a potency similar to that of 1 but has much improved solubility over 1. We then made four shorter peptides (peptides 3−6) by truncating one residue at a time from either the C- or Nterminus of peptide 2 (Table 1). Peptide 3 with a serine residue deleted from the C-terminus in peptide 2 binds to RAP1 with Ki = 5.8 μM, and is 3 times weaker than peptide 2. Removal of the C-terminal Leu295 in peptide 3 yielded peptide 4, which has no discernible binding to RAP1 at 200 μM, revealing the critical contribution of Leu295 to the binding. Peptide 5 with Thr281 residue removed from the N-terminus in peptide 3 binds to RAP1 with Ki = 19.5 μM, and is 3 times weaker than peptide 3. Further removal of the N-terminal Thr282 residue in peptide 5 generated peptide 6, which reduced the binding affinity to RAP1 by an additional factor of 2. Upon the basis of these data, we decided to employ the 16-mer peptides 1 and 2 for the design and synthesis of triazole-stapled peptides. Identification of Stapling Sites. For the design of triazole-stapled peptides, it is necessary first to identify appropriate stapling sites. The cocrystal structure of TRF2 complexed with RAP1 shows that six residues (Met285, Met286, Lys289, Ala290, Lys293, and The294) on the α1 helix are exposed to solvent (Figure 2). We mutated Met285, Met286, Ala290, and Thr294 individually to lysine, yielding peptides 7, 2, 8, and 9, respectively (Table 2), in order to determine if any of these sites are suitable for stapling. Binding data showed that with the exception of mutation of Met285 to lysine, changing any of the

protein interaction and disrupting the entire shelterin complex. Such compounds may prove to be powerful tools with which to investigate the fundamental role of shelterin in regulation of telomeres and may have potential for the treatment of human diseases. To date, compounds that target the RAP1/TRF2 protein−protein interaction have not been reported, and in this study, we describe our efforts in the design and development of the first, high-affinity stapled peptides targeting the TRF2/ RAP1 protein−protein interaction. In recent years, stapled peptides have been developed as an alternative to non-peptide, small-molecule inhibitors targeting protein−protein interactions in cases in which those interactions are mediated by a helical segment from one binding partner.12,13 Perhaps the most noteworthy stapling method is the ring-closing metathesis, optimized by Verdine and colleagues.12−14 Research using the ring-closing metathesis method has generated high-affinity and cell-permeable stapled peptides that interfere with a number of protein−protein interactions.12−14 In addition to the ring-closing metathesis stapling method, we have developed a triazole-stapling strategy which has been applied successfully to the design of BCL9 stapled peptides which block the BCL9/β-catenin protein− protein interaction.15 The triazole-stapling method employs a Huisgen 1,3-dipolar cycloaddition reaction, which is efficient under mild conditions,16,17 and the stapling residues employed either are commercially available or can be readily prepared.18 In our design of stapled TRF2 peptides, we employed the triazole-stapling method because the wild-type TRF2 peptide has a very poor solubility and the ring-closing metathesis method, using an all-hydrocarbon staple, would lead to stapled peptides with even poorer solubility. Our study has yielded a number of high affinity stapled peptides, the most potent of which binds to RAP1 with a Ki value of 7 nM and is 300 times more potent than the initial unstapled TRF2 peptide.

■

19.5 ± 0.9 39.1 ± 0.4

RESULTS

Determination of Minimum Sequence of TRF2 for Binding to RAP1. Although the 41-residue TRF2 RBM domain binds to RAP1 RCT with a high affinity (Kd = 16.5 nM),11 it contains two helical segments connected by four unstructured residues and is not suitable for the design of stapled peptides. As the α1 helical segment of TRF2 RBM domain appears to contribute significantly to its interaction with RAP1, we reduced the 41-mer RBM domain to a 16-mer peptide, consisting of the 12-mer α1 helix and the unstructured 4-residue loop at its N-terminus (Table 1). This peptide (1, Table 1) binds to RAP1 protein with Ki = 2.0 μM, and its further truncation at either the N- or the C-terminus resulted in peptides with very poor solubility, which prevented accurate determination of their binding affinities in our FP assay. Consequently, we decided to use peptide 1 as the starting point for further modifications.

Figure 2. Binding mode of 16-residure peptide 1 (yellow) to RAP1 (surface). Solvent exposed residues are shown as sticks. 329

DOI: 10.1021/acs.jmedchem.5b01465 J. Med. Chem. 2016, 59, 328−334

Journal of Medicinal Chemistry

Article

Table 2. Binding Affinity of Peptides with a Lysine Mutation on Solvent Exposed Residues, Determined in Our Optimized FP Competitive Binding Assay 1 7 2 8 9

sequence

IC50 (μM)

Ki (μM)

Ac-281TTIGMMTLKAAFKTLS296-NH2 Ac-281TTIGKMTLKAAFKTLS296-NH2 Ac-281TTIGMKTLKAAFKTLS296-NH2 Ac-281TTIGMMTLKKAFKTLS296-NH2 Ac-281TTIGMMTLKAAFKKLS296-NH2

9.3 ± 0.4 64.4 ± 8.8 8.3 ± 0.4 13.8 ± 0.6 7.1 ± 0.4

2.0 ± 0.1 13.9 ± 1.9 1.8 ± 0.1 3.0 ± 0.1 1.6 ± 0.1

other three residues to lysine had little or no effect on binding of the peptide to RAP1. The data thus suggested that Met286, Ala290, Lys289, Lys293, and Thr294 could be used as stapling sites in our design of triazole-stapled peptides. Although Lys289 and Lys293 are suitable for stapling, we found that their presence is important because it maintains adequate solubility for peptides 1 and 2, and we decided to employ three other residues (Met286, Ala290, and Thr294) for the design of stapled peptides. Design of Triazole-Stapled Peptides. Using Met286 and Ala290 as stapling sites, we replaced the side chain of Met286 with an azide and the side chain of Ala290 with an alkyne for triazole stapling. Upon the basis of our previous study,15 we elected to create a stapled structure with a seven-membered ring system having three-additional backbone amide bonds. For the alkyne at residue 290, we tested both the L- and Dconfigurations, since our previous study showed that using a Dconfigured alkyne residue at i + 4 position yields stapled peptides with much better affinities than using a L-configuration alkyne residue.15 Using Ala290 and Thr294 as the stapling sites, we replaced the side chain of Ala290 with an azide with Lconfiguration and the side chain of Thr294 with an alkyne with either L- or D-configuration. For comparisons, we synthesized both stapled and corresponding unstapled peptides (Figure 3).

Table 3. Binding Affinity of Linear and Stapled Peptides, Determined in Our Optimized FP Competitive Binding Assay IC50 (μM) 10 11 12 13 14 15 16 17 a

137.1 43.3 37.8 36.5 >15a 0.7 ± 0.1 9.7 ± 0.6 28.5 ± 14.2

Ki (μM) 50.2 15.7 13.7 13.2 0.14 ± 0.03 2.1 ± 0.1 6.2 ± 3.0

Precipitated at 15 μM.

affinities as compared to peptides 1 and 2. Formation of a triazole group between the azide and alkyne groups in 10 yielded the stapled peptide 11 but improved the binding affinity to RAP1 by a factor of only 3 over that of 10. In contrast, formation of a triazole group between the azide and alkyne groups in 14 yielded the stapled peptide 15, which has a Ki value of 0.14 μM and is thus >20 times more potent than 14 and >10 times more potent than the original peptide 1. Furthermore, while 15 is completely soluble in phosphate buffer at a concentration of 100 μM, the unstapled peptide 14 precipitates at 15 μM. These data show that stapling not only enhances the binding affinity to RAP1 but also improves the compound’s solubility over that of its unstapled counterpart. The unstapled peptide 12 with two L-amino acids at the stapling sites (residues 286 and 290) is >5 times less potent than peptide 1, but the unstapled peptide 16 with two L-amino acids at the stapling sites (residues 290 and 294) has an affinity very similar to that of peptide 1. However, stapling both 12 and 16 failed to improve their binding to RAP1 (12 vs 13 and 16 vs 17), and in fact, the stapled peptide 17 is 3 times less potent in binding to RAP1 than the corresponding unstapled peptide 16. Taken together, our data show that the stapling strategy in stapled peptide 15 significantly improves its binding affinity to RAP1 over its unstapled counterpart 14. Improvement of Binding Affinity through Mutations of Ile283. The stapled peptide 15 binds to RAP1 with a Ki value of 0.14 μM and is a promising lead peptide. We performed additional modifications with the goal of further improving its binding affinity to RAP1. The cocrystal structure of RAP1 in a complex with TRF2 shows that the side chain of Ile283 interacts with a large hydrophobic pocket in RAP1. Our modeling analysis suggested that a larger hydrophobic residue than Ile at this site should be tolerated and may lead to compounds with high affinities. We have therefore synthesized a number of stapled peptides in which we replaced Ile283 with either natural or unnatural hydrophobic amino acids. The binding data for these stapled peptides to RAP1 are provided in Table 4.

Figure 3. Structures of both linear (10, 12, 14, and 16) and stapled peptides (11, 13, 15, and 17). ∗ denotes L amino acid, and ∧ denotes D amino acid.

Binding Affinities of Triazole-Stapled Peptides. We tested binding of these stapled and unstapled peptides to RAP1 using our optimized FP binding assay and the data are summarized in Table 3. As expected, the unstapled peptides 10 and 14 with one Land one D-amino acid at the stapling sites have reduced binding 330

DOI: 10.1021/acs.jmedchem.5b01465 J. Med. Chem. 2016, 59, 328−334

Journal of Medicinal Chemistry

Article

content of 37% (Figure 4), we could not obtain a reliable CD spectrum for unstapled peptide 14 due to its poor solubility.

Table 4. Binding Affinity of Ile283 Mutated Stapled Peptides to RAP1 Protein, Determined in Our Optimized FP Competitive Binding Assaya

Figure 4. Circular dichroism spectra of stapled peptides 11, 15, and 20 and unstapled peptide 26, which is the counterpart of stapled peptide 20. ∗ and ∧ denote L and D amino acid, respectively.

a

∗ and ∧ denote

L

and

D

To circumvent the poor solubility problem observed in the case of peptide 14, we synthesized the corresponding unstapled peptide 26 based upon stapled peptide 20. The unstapled peptide 26 was found to have a much better solubility than 14. Our binding assay showed that 26 has a Ki value of 18 μM and is nearly 1000 times less potent than its stapled counterpart 20 (Ki = 19 nM). CD spectrometry showed that while the unstapled peptide 26 adopts a random coil conformation in phosphate buffer, the stapled peptide 20 has a helical content of approximately 30% (Figure 4). Stapling therefore greatly improves the helical propensity of the corresponding unstapled peptide. To examine if the 100-fold binding affinity difference between stapled peptides 11 and 15 to RAP1 is attributed to their different helical contents, we determined the CD spectrum for 11, which showed that 11 has a random coiled structure (Figure 4). Hence, the much higher affinity of 15 than 11 is attributed to, at least in part, the much higher helical propensity for 15. Design and Optimization of Fluorescent Probes for Fluorescence-Polarization (FP) Binding Assays. In order to determine the binding affinities of our designed peptides to RAP1, we have developed and optimized sensitive and accurate competitive fluorescence polarization (FP) based binding assays. We first designed and synthesized two fluorescent probes (27 and 28) by tethering 5-carboxyfluorescein (5-FAM) to either the N- or C-terminal of peptide 1, with a spacer (Figure 5). 5-FAM is one of the most widely used fluorophores for its high quantum yield and well-studied chemistry. A flexible spacer was utilized in the design of 27 and 28 to minimize the influence of 5-FAM to the interaction between peptide 1 and RAP1. Protein titration experiments were performed to determine the Kd values of 27 and 28 to RAP1. Probe 28 with the 5-FAM tethered to the C-terminus binds to RAP1 with a Kd value of 0.3 μM, whereas probe 27 with 5-FAM tethered to the N-terminus has a Kd value of 2.2 μM (Figure 6). Both probes 27 and 28 produced dynamic ranges larger than 100

amino acid, respectively.

The stapled peptide 18 with a 2,2-dimethylpropyl side chain in residue 283 has a Ki value of 53 nM to RAP1 and is thus 3 times more potent than 15. Compound 19 with a methylcyclohexyl side chain at the same position has a Ki value of 32 nM and is 4 times more potent than 15. Replacing Ile283 with a Phe residue yielded 20, which binds to RAP1 with a Ki value of 19 nM and is 7 times more potent than 15. Replacing Ile283 with a Trp residue, which led to 21, did not show improved binding affinity over that of 15 (0.13 μM vs 0.14 μM), but the stapled peptide 22, with a 2-methylnaphthalenyl side chain at position 283 with a Ki value of 16 nM, is 8 times more potent than 15. Both stapled peptide 22 with a 2-methylnaphthalenyl side chain and stapled peptide 20 with a methylphenyl side chain at residue 283 have high binding affinities to RAP1. We next investigated if a chorine substituent on the phenyl ring in 20 could further enhance its binding affinity to RAP1. While substitution by chlorine at either the meta- or para-position does not affect the binding affinity to RAP1 of the resulting compounds 24 and 25, ortho-chloro substitution of the phenyl ring, as in compound 23, improves the binding affinity by a factor of 2. Compound 23 has a Ki value of 7 nM to RAP1 and is the most potent compound obtained in this study. Effect of Stapling on Helical Propensity. We investigated if the improvement in the binding affinity of stapled peptide 15 over its unstapled counterpart 14 can be attributed to enhancement of the secondary helical structure using circular dichroism (CD) spectrometry. While stapled 15 has a helical 331

DOI: 10.1021/acs.jmedchem.5b01465 J. Med. Chem. 2016, 59, 328−334

Journal of Medicinal Chemistry

Article

Figure 5. Structures of fluorescently tagged probes designed based upon linear peptide 1 and stapled peptide 20.

ΔmP, which is excellent for developing robust competitive assays. Since 28 has a much higher affinity to RAP1 than 27, 28 was employed for the development of our initial competitive FP binding assay. In the initial competitive assay, 600 nM RAP1 protein and 2 nM probe 28 were used. Additionally, 4% DMSO was introduced in the assay mixture in order to enhance the solubility of our designed peptides in the aqueous assay buffer. With more potent compounds obtained whose Ki values could not be accurately determined with the assay using probe 28, we designed and synthesized two second generation fluorescent probes 29 and 30 by tethering 5-FAM to either the N- or C-terminus of the more potent peptide 20,

Figure 6. Saturation curves of RAP1 protein to four different fluorescently tagged tracers.

Figure 7. (A). Competitive binding curves of representative peptides to RAP1 using the initial fluorescence probe 28. (B) Competitive binding curves of representative peptides to RAP1 using a more potent fluorescence probe 29. 332

DOI: 10.1021/acs.jmedchem.5b01465 J. Med. Chem. 2016, 59, 328−334

Journal of Medicinal Chemistry

Article

concentrations up to full saturation. Fluorescence polarization values were measured using the Infinite M-1000 plate reader (Tecan U.S., Research Triangle Park, NC) in Microfluor 1 96-well, black, roundbottom plates (Thermo Scientific). Serial dilutions of the protein being tested were mixed with the tracer to a final volume of 125 μL in the assay buffer (100 mM potassium phosphate, pH 7.5, 100 μg/mL bovine γ-globulin, 0.02% sodium azide, Invitrogen, with 0.01% Triton X-100 and 4% DMSO). The final tracer concentration was 2 nM. Plates were incubated at room temperature for 1−2 h with gentle shaking to ensure equilibrium. The polarization values in millipolarization units (mP) were measured at an excitation wavelength of 485 nm and an emission wavelength of 530 nm. Equilibrium dissociation constants (Kd) were then calculated by fitting the sigmoidal dosedependent FP increases as a function of protein concentrations using GraphPad Prism 5.0 software (GraphPad Software, San Diego, CA). The influence of DMSO, detergent, and incubation time was evaluated to determine the optimal conditions under which the Kd value showed minimal variation. The IC50 and Ki values of compounds were determined through a compound dose-dependent competitive binding experiment in which serial dilutions of compounds competed against a fixed concentration of the fluorescent probe for binding to the protein with a fixed concentration (typically 2−3 times the Kd values determined above). Mixtures of 5 μL of the tested compounds in DMSO and 78 μL of preincubated protein/probe complex solution in the assay buffer (100 mM potassium phosphate, pH 7.5, 100 μg/mL bovine γ-globulin, 0.02% sodium azide, Invitrogen with 0.01% Triton X-100) were added into assay plates and incubated at room temperature for 1 h with gentle shaking. Final concentrations of the protein and probe were 40 nM and 2 nM, respectively. Negative controls containing protein/ probe complex only (equivalent to 0% inhibition) and positive controls containing only free probes (equivalent to 100% inhibition) were included in each assay plate. FP values were measured as described above. IC50 values were determined by nonlinear regression fitting of the competition curves. The Ki values of competitive inhibitors were calculated using the derived equation described previously, based upon the measured IC50 values, the Kd value of the probe to the protein, and the concentrations of the protein probes in the competitive assays.19,20 Circular Dichroism. Circular dichroism experiments were performed using a Jasco J715 spectropolarimeter. Peptides were dissolved in 10 mM phosphate buffer (pH = 7.4) to produce a ∼100 μM solution. Spectrum generation and percentage helicity calculations were performed using the method described in the previous study.15,21

respectively, with the same spacers as used in the synthesis of probes 27 and 28, respectively. Saturation experiments determined that probes 29 and 30 have the same Kd value of 11 nM to RAP1 (Figure 6) and produce dynamic ranges of 81 and 89 ΔmP, respectively, sufficiently large for the development of an accurate, competitive binding assay. Employing probe 29, we reoptimized the binding assay conditions and retested all synthesized peptides for their binding affinities to RAP1, and all the data reported in this study were obtained using this optimized competitive FP assay. With probe 29 in hand, we were able to reduce the protein concentration in the competitive assay to as low as 40 nM but still achieved dynamic range greater than 50 which is sufficient to obtain a good signal-to-noise ratio. To illustrate the advantage using the higher affinity probe 29 over lower affinity probe 28, Figure 7 depicts competitive binding curves obtained using probes 28 and 29 for several representative compounds with different affinities. As can be seen, while the competitive binding assay using probe 28 can clearly distinguish the binding affinity difference among compounds 8, 9, and 15, it is unable to distinguish the binding affinity difference between two high-affinity compounds 15 and 18 (Figure 7A). In comparison, the competitive binding assay using probe 29 can clearly distinguish the binding affinity difference among compounds 8, 9, 15, and 18 (Figure 7B).

■

SUMMARY In this study, we have designed and synthesized the first TRF2based stapled peptides targeting protein−protein interactions between the RAP1 and TRF2 proteins in the shelterin complex. Through truncation, triazole stapling, and mutations of Ile283, we obtained a number of stapled peptides with high binding affinity to RAP1. The most potent peptide (23) binds to RAP1 with Ki value of 7 nM and is nearly 300 times more potent than the corresponding wild-type TRF2 peptide 1 (Ki = 2 μM). Upon the basis of the high affinity stapled peptide 20, we have designed two fluorescently tagged probes (29 and 30), and using probe 29, we have developed and optimized a sensitive, quantitative and competitive FP-based binding assay. This assay was used for accurate determination of the binding affinities of all of our designed peptides and could be also employed for high throughput screening of non-peptide small-molecule inhibitors. Further modifications of these stapled peptides may yield a set of potent and cell-permeable stapled peptides as pharmacological tools with which to investigate the roles of the TRF2/RAP1 protein−protein interaction and the shelterin complex in regulation of telomeres and to validate the TRF2/ RAP1 protein−protein interaction as a potential therapeutic target.

■

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jmedchem.5b01465. Purity information for all of the final compounds (PDF) Molecular strings for all the compounds reported in this study and their associated biochemical binding data (CSV)

■

EXPERIMENT

Peptide Synthesis. All the peptides were synthesized using Fmoc solid phase peptide synthesis (SPPS) strategy using an ABI 433A peptide synthesizer. The conditions regarding synthesis and purification followed previously reported methods.15 Fluorescence Polarization (FP) Assay. Competitive FP binding assays were designed and optimized to determine the binding affinities of synthesized peptides to the RAP1 protein. A fluorescent probe 29 was designed and synthesized by tethering FAM to the N terminus of a stapled TRF2 peptide 20 which binds to RAP1 with high affinity, as determined by preliminary experiments. The Kd value of the tracer with the RAP1 protein was determined by monitoring the total fluorescence polarization of mixtures with the fluorescent probe at a fixed concentration and proteins at increasing

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Phone: (734) 615-0362. Notes

The authors declare no competing financial interest.

■

ABBREVIATIONS USED TRF1, telomeric repeat-binding factor 1; TRF2, telomeric repeat-binding factor 2; TIN2, telomeric repeat-binding factor 1 interacting nuclear factor 2; RAP1, repressor activator protein 1; TPP1, tripeptidyl peptidase 1; POT1, protection of telomere 333

DOI: 10.1021/acs.jmedchem.5b01465 J. Med. Chem. 2016, 59, 328−334

Journal of Medicinal Chemistry

Article

(21) Luo, P. Z.; Baldwin, R. L. Mechanism of helix induction by trifluoroethanol: a framework for extrapolating the helix-forming properties of peptides from trifluoroethanol/water mixtures back to water. Biochemistry 1997, 36, 8413−8421.

1; RBM, repressor activator protein 1 binding motif; RCT, repressor activator protein 1 C-terminal domain; FP, fluorescence polarization; BCL9, B-cell CLL/lymphoma 9 protein; CD, circular dichroism

■

REFERENCES

(1) Janouskova, E.; Necasova, I.; Pavlouskova, J.; Zimmermann, M.; Hluchy, M.; Marini, V.; Novakova, M.; Hofr, C. Human Rap1 modulates TRF2 attraction to telomeric DNA. Nucleic Acids Res. 2015, 43, 2691−2700. (2) Sarthy, J.; Bae, N. S.; Scrafford, J.; Baumann, P. Human RAP1 inhibits non-homologous end joining at telomeres. EMBO J. 2009, 28, 3390−3399. (3) Bombarde, O.; Boby, C.; Gomez, D.; Frit, P.; Giraud-Panis, M. J.; Gilson, E.; Salles, B.; Calsou, P. TRF2/RAP1 and DNA-PK mediate a double protection against joining at telomeric ends. EMBO J. 2010, 29, 1573−1584. (4) Sfeir, A.; Kabir, S.; van Overbeek, M.; Celli, G. B.; de Lange, T. Loss of Rap1 induces telomere recombination in the absence of NHEJ or a DNA damage signal. Science 2010, 327, 1657−1661. (5) Bae, N. S.; Baumann, P. A RAP1/TRF2 complex inhibits nonhomologous end-joining at human telomeric DNA ends. Mol. Cell 2007, 26, 323−334. (6) Celli, G. B.; de Lange, T. DNA processing is not required for ATM-mediated telomere damage response after TRF2 deletion. Nat. Cell Biol. 2005, 7, 712−718. (7) Li, B.; Oestreich, S.; de Lange, T. Identification of human Rap1: implications for telomere evolution. Cell 2000, 101, 471−483. (8) Li, B.; de Lange, T. Rap1 affects the length and heterogeneity of human telomeres. Mol. Biol. Cell 2003, 14, 5060−5068. (9) O’Connor, M. S.; Safari, A.; Liu, D.; Qin, J.; Songyang, Z. The human Rap1 protein complex and modulation of telomere length. J. Biol. Chem. 2004, 279, 28585−28591. (10) de Lange, T. Shelterin: the protein complex that shapes and safeguards human telomeres. Genes Dev. 2005, 19, 2100−2110. (11) Chen, Y.; Rai, R.; Zhou, Z. R.; Kanoh, J.; Ribeyre, C.; Yang, Y.; Zheng, H.; Damay, P.; Wang, F.; Tsujii, H.; Hiraoka, Y.; Shore, D.; Hu, H. Y.; Chang, S.; Lei, M. A conserved motif within RAP1 has diversified roles in telomere protection and regulation in different organisms. Nat. Struct. Mol. Biol. 2011, 18, 213−221. (12) Verdine, G. Drugging “undruggable” targets using stapled peptides. Biopolymers 2009, 92, 296−296. (13) Schafmeister, C. E.; Po, J.; Verdine, G. L. An all-hydrocarbon cross-linking system for enhancing the helicity and metabolic stability of peptides. J. Am. Chem. Soc. 2000, 122, 5891−5892. (14) Barthe, P.; Rochette, S.; Vita, C.; Roumestand, C. Synthesis and NMR solution structure of an alpha-helical hairpin stapled with two disulfide bridges. Protein Sci. 2000, 9, 942−955. (15) Kawamoto, S. A.; Coleska, A.; Ran, X.; Yi, H.; Yang, C. Y.; Wang, S. Design of triazole-stapled BCL9 alpha-helical peptides to target the beta-catenin/B-cell CLL/lymphoma 9 (BCL9) proteinprotein interaction. J. Med. Chem. 2012, 55, 1137−1146. (16) Huisgen, R. Centenary lecture1,3-Dipolar cycloadditions. Proc. Chem. Soc., London 1961, 357−396. (17) Kolb, H. C.; Finn, M. G.; Sharpless, K. B. Click chemistry: Diverse chemical function from a few good reactions. Angew. Chem., Int. Ed. 2001, 40, 2004−2021. (18) Katayama, H.; Hojo, H.; Ohira, T.; Nakahara, Y. An efficient peptide ligation using azido-protected peptides via the thioester method. Tetrahedron Lett. 2008, 49, 5492−5494. (19) Huang, X. Y. Fluorescence polarization competition assay: The range of resolvable inhibitor potency is limited by the affinity of the fluorescent ligand. J. Biomol. Screening 2003, 8, 34−38. (20) Nikolovska-Coleska, Z.; Wang, R. X.; Fang, X. L.; Pan, H. G.; Tomita, Y.; Li, P.; Roller, P. P.; Krajewski, K.; Saito, N. G.; Stuckey, J. A.; Wang, S. M. Development and optimization of a binding assay for the XIAP BIR3 domain using fluorescence polarization. Anal. Biochem. 2004, 332, 261−273. 334

DOI: 10.1021/acs.jmedchem.5b01465 J. Med. Chem. 2016, 59, 328−334