An evolution-inspired strategy to design disulfide-rich peptides tolerant to extensive sequence manipulation

Natural disulfide-rich peptides (DRPs) are valuable scaffolds for the development of new bioactive molecules and therapeutics. However, there are only a limited number of topologically distinct DRP folds in nature, and most of them suffer from the problem of in vitro oxidative folding. Thus, strategies to design DRPs with new constrained topologies beyond the scope of natural folds are desired. Herein we report a general evolution-inspired strategy to design new DRPs with diverse disulfide frameworks, which relies on the incorporation of two cysteine residues and a random peptide sequence into a precursor disulfide-stabilized fold. These peptides can spontaneously fold in redox buffers to the expected tricyclic topologies with high yields. Moreover, we demonstrated that these DRPs can be used as templates for the construction of phage-displayed peptide libraries, enabling the discovery of new DRP ligands from fully randomized sequences. This study thus paves the way for the development of new DRP ligands and therapeutics with structures not derived from natural DRPs.


Materials and instruments
All the Fmoc-protected amino acids and Rink amide MBHA resin used for peptide synthesis were supplied by GL Biochem (Shanghai, China). Glutathione oxidized (GSSG), glutathione reduced (GSH) and tris (2-carboxyethyl)

S4
All peptides were synthesized using solid-phase peptide synthesis by a CEM Discover Liberty BLUE microwave-assisted peptide synthesizer. High performance liquid chromatography (HPLC, SHIMADZU) equipped with a quaternary solvent manager (QSM), an AQUITY ® PDA detector and a column manager was used for analysis of peptides. Reaction progress was monitored by HPLC (a flow rate of 1.0 mL· min -1 flow rate of H2O (+0.1% TFA) and ACN (+0.1% TFA); isocratic with 10% ACN (+0.1% TFA) for 5.0 min followed by a linear gradient of 10% to 85% ACN (+0.1% TFA) over 30 min). A HITACHI U-3900H UV/Vis spectrometer was used for the quantification of peptides. Surface Plasmon Resonance (SPR) assays were performed by Biacore T200.  Hitrap TM Desalting column (GE Healthcare) was used to purify the biotinylated proteins.

Synthesis of peptides
All peptides are N-terminally acetylated and C-terminal amidated, which were synthesized at 0.05 mmol scale using the Fmoc solid-phase peptide synthesis (SPPS) on a CEM Liberty blue automated microwave peptide synthesizer (Table S1). Amino acids were coupled onto the MBHA resins using the standard coupling protocol. Peptides were cleaved from the resin and deprotected by treating with a TFA cleavage cocktail for 4 h at room temperature. Then, the cleaved peptides were precipitated in cold diethyl ether, and purified using a HPLC system. All peptides were purified to a purity of >95% before performing the oxidation (Figures S1 and S7).

Oxidation of peptides
Reduced peptides (>95% purity) isolated by HPLC were identified by mass spectrometry (Table   S2). The HPLC-isolated peptides were lyophilized and re-dissolved in a mixture of water and acetonitrile (70%:30% vol/vol) as the stock solution for further use. Concentrations of the reduced S5 peptide stock solutions were determined by UV/Vis spectroscopy (ɛTrp = 5,502 cm -1 M -1 at 280 nm).
In a typical experiment for peptide oxidation, the reduced peptide was reconstituted in an appropriate amount of phosphate buffer (100 mM, pH 7.4) containing 30% acetonitrile, 0.5 mM GSH and 0.5 mM GSSG to achieve a concentration of 50 μM. After about 4-6 h, the oxidation was complete, and the oxidized peptide was characterized by HPLC and mass spectrometry (Table S3). The Acm-protected peptides were oxidized in the redox buffer in the same way. After the HPLC-purification of the oxidized Acm-protected peptides, the peptide was lyophilized and reconstituted in pure water to achieve a definite concentration (determined by UV/Vis spectroscopy).
Then, the peptide solution was diluted by addition of methanol with 0.1% TFA to a concentration of 50 μM, into which 10 eq. I dissolved in methanol was added dropwise. After ~1 h, the Acm was completely removed from the peptide, leading to the formation of the third disulfide bond in the peptide. The final Acm-deprotected product was analyzed using HPLC and mass spectrometry (Table S4). To characterize the disulfide pairing of oxidized 12 and 14, the lyophilized powder (oxidized 12 and 14 isolated by HPLC) was dissolved in a phosphate buffer (100 mM, pH 6.0) containing 50 μg/mL trypsin for 4 h, and then the major digested peptide fragment isolated by HPLC (monitored at 280 nm) was identified using mass spectrometry ( Figures S9 and S10).

Construction of phage-displayed peptide libraries
The peptide libraries were constructed based on a procedure reported previously (Table S5). 1 In brief, DNA libraries encoding random peptide sequences were digested by Sfi I (10 h, 50 °C ) and Not I (10 h, 37 °C ), which were recovered by gel purification. Then, the purified DNA fragments were S6 ligated with gel-purified Sfi I/Not I-digested vector pCantab 5E (ratio of insert and vector: 10/1). The ligation mixture was then transformed into E. coli TG1 competent cells. The cells were plated on 2 × YT/ampicillin agar plates and incubated at 37 °C for 12 h. The size of the phage libraries was determined by measuring the total number of colonies. The colonies on the plates were then scraped off the plates and propagated for phage production and purification. 20-30 phage clones were randomly picked up for sequencing to evaluate the quality of the phage libraries.

Phage panning
Proteins (MDM2 and Bcl-2; 5 μM) were biotinylated in phosphate buffer saline (PBS; pH 7.4) with the addition of Sulfo-NHS-LC-biotin (50 μM) for ~0.5 h at room temperature. After the reaction, the unreacted Sulfo-NHS-LC-biotin was removed using a desalt column on an AKTA pure system (running buffer: pH 7.4 PBS). The biotinylated proteins can then be immobilized on streptavidin-coated and neutravidin-coated magnetic beads for screening as described previously. 2,3 Briefly, 100 μL of streptavidin-coated magnetic beads was washed three times with binding buffer (150 mM NaCl, 1 mM CaCl2, 10 mM Tris-HCl, 10 mM MgCl2, pH 7.4) in a 1.5 mL microcentrifuge tube, which were then re-suspended with 100 μL binding buffer and distributed equally into two 1.5 mL microcentrifuge tubes. Then, the biotinylated protein (first round: 10 μg; second round: 5 μg; third round: 2 μg) was added into one of the two tubes; meanwhile, the same volume of 1 × PBS was added into the other one as a control. The two tubes were incubated on a slowly rotating vortex mixer for 15 min at room temperature. Then, the beads in the two tubes were washed three times with the binding buffer to remove the free proteins, which were then resuspended with 300 μL S7 binding buffer and 150 μL Blocking buffer (binding buffer containing 0.3% Tween-20 and 3% w/v BSA). The suspension of beads was incubated at room temperature for 2 h on a slowly rotating vortex mixer. Meanwhile, to a tube containing 3.0 mL of phages (>10 12 t.u.) dissolved in binding buffer, 1.5 mL blocking buffer was added, and the solution was incubated at room temperature for 2 h on a slowly rotating vortex mixer. Then, the phages in solution were split equally into two tubes, into which the BSA-blocked beads with and without the immobilized proteins were added, respectively. After 30 min incubation at room temperature, the unbound phages in the supernatant were removed and the beads were washed for nine times with washing buffer (binding buffer containing 0.1% Tween-20) and twice with binding buffer. During the washing steps, the tubes were replaced at least three time with new ones to avoid the non-specific adsorption of phages. After the last washing step, the beads were resuspended with 200 μL elution buffer (50 mM glycine, pH 2.2) and incubated at room temperature for 5 min. Then, the beads were magnetically precipitated and the supernatant was transferred to a new microcentrifuge tube containing 50 μL neutralization buffer (1 M Tris-Cl, pH 8.9). The eluted phages from both the experimental and control group were diluted to infect exponentially growing TG1 cells to quantify the phage titer, and the eluted phages from the experimental group were propagated for the production of phages used for next-round panning. After three rounds of panning as described above, 20-30 phage clones were randomly picked for sequencing. Note that streptavidin-coated and neutravidin-coated magnetic beads were used to immobilize biotinylated proteins alternatively in each round of selection to prevent the enrichment of streptavidin and neutravidin binders during the panning. 2,4 For the library screening against streptavidin, streptavidin-coated magnetic beads were directly used for panning.

Fluorescence polarization (FP) assays
FP assays were performed on 96-well flat-bottom OptiPlate black plates using a Tecan Infinite® 200 PRO Microplate Reader. To examine the binding affinity of oxidized 16 to Bcl-2, a fluorescein-labelled 16 (FITC-ꞵAla-16) was synthesized and purified to a purity of >95% using HPLC. Fluorescein-labelled 16 was first dissolved in DMSO and then diluted with PBS (pH 7.4) for further use. Sumo-Bcl-2 was obtained from an E. coli cell expression system described previously. 5 For the FP assay, 20 nM of FITC-ꞵAla-16 was incubated with Sumo-Bcl-2 (0-1.2 μM) in PBS (pH 7.4) on a 96-well plate at room temperature for 10 min, and fluorescence anisotropies were recorded.
The KD value of the binding between FITC-ꞵAla-16 and Bcl-2 can then be obtained by fitting the data with the single-site binding model. For the FP competition assay, 20 nM of FITC-ꞵAla-16 and 400 nM of Sumo-Bcl-2 were incubated with oxidized 16 (0-10 μM) under the same conditions as described above, and fluorescence anisotropies were recorded. The Ki value of the binding between oxidized 16 and Bcl-2 can then be obtained using a data-fitting procedure reported previously. 5 The binding of peptides to MDM2 was evaluated using the same procedure described above, which has been described in our previous reports. 1,6

Surface plasmon resonance (SPR) assays
SPR assays were performed using a Biacore T200. Target proteins (Sumo-Bcl-2 and streptavidin) were immobilized on chip surface using a procedure described previously. 5 Briefly, Flow cell Fc4 was activated first with a mixture of NHS and EDC, and then Sumo-Bcl-2 (or streptavidin) dissolved in 10 mM acetate buffer (pH 5.0) flowed through the cell for protein coupling S9 to reach a target response of 1000 RU. Flow cell Fc3 that was not immobilized with proteins was used as the reference channel. Serially diluted samples were then passed over the flow cells at a flow rate of 30 μL/min (160, 320, 640, 1280, 2560, 640 nM for oxidized 15, 16, and 17; 20, 40, 80, 160, 320, 80 nM for oxidized 11, 15, and 16; 5, 10, 20, 40, 80 nM for oxidized 17). Kinetics data were then analyzed using a 1:1 binding model and local fit to obtain kinetic rate constant and dissociation constant.

NMR characterization
NMR samples containing 0.5 mM of oxidized 15 and oxidized 16 were prepared in 50% perdeuterated acetonitrile and 50% H2O. NMR experiments were recorded at 298 K on Bruker AVANCE III 600 MHz and 850 MHz equipped with a cryogenic triple-resonance probe for oxidized 15 and 16, respectively. Two dimensional (2D) 1 H, 15 N/ 13 C HSQC spectra were recorded to obtain chemical shifts of 13 C/ 15 N in backbone and side-chains. 2D 1 H-1 H TOCSY (80 ms) and 2D 1 H-1 H COSY spectra were acquired to assign NMR signals of the peptide, and 2D 1 H-1 H NOESY spectra with a mixing time of 300 ms were measured to obtain 1 H-1 H distance constraints. The NMR data were processed using NMRPipe/NMRDraw and analyzed using NMRFAM-SPARKY. 7 Figure S1. Chromatograms of peptides 1-8.