Monomer Choice Influences N-Acryloyl Amino Acid Grafter Conversion via Protease-Catalysis

Abstract

End-capped peptides modified with reactive functional groups on the N-terminus provide a route to prepare peptide-polymer conjugates for a broad range of applications. Unfortunately, current chemical methods to construct modified peptides rely largely on solid-phase peptide synthesis (SPPS), which lacks green preparative characteristics and is costly, thus limiting its applicability to specialty applications such as regenerative medicine. This work evaluates N-terminally modified N-acryloyl-glutamic acid diethyl ester, N-acryloyl-leucine ethyl ester, and N-acryloyl-alanine ethyl ester as grafters, and papain as the protease, for the direct addition of amino acid ethyl ester (AA-OEt) monomers via protease catalyzed peptide synthesis (PCPS) and the corresponding formation of N-acryloyl functionalized oligopeptides in a one-pot aqueous reaction. It was hypothesized that, by building N-acryloyl grafters from AA-OEt monomers that are known to be good substrates for papain in PCPS, the corresponding grafters would yield high grafter conversions, high ratio of grafter-oligopeptide to free NH₂-oligopeptide, and high overall yield. However, this work demonstrates based on the grafter/monomers studied herein that the dominant factor in N-acryloyl-AA-OEt grafter conversion is the co-monomer used in co-oligomerizations. Computational modeling using Rosetta qualitatively recapitulates the results and provides insight into the structural and energetic bases underlying substrate selectivity. The findings herein expand our knowledge of factors that determine the efficiency of preparing N-acryloyl terminated oligopeptides by PCPS that could provide practical routes to peptide macromers for conjugation to polymers and surfaces for a broad range of applications.

Graphical Abstract

INTRODUCTION

Peptide conjugates are a versatile family of hybrid molecules tailorable to meet performance needs in numerous applications, including regenerative medicine.¹ The conjugation of peptides to polymers,² surfaces,³ or nanoparticles⁴ combines the biological and functional properties built into peptides with the physical characteristics of the other conjugate moiety. Examples of properties peptides can provide to hybrid molecules include, but are not limited to, self-assembly,⁵ metal chelation,⁶ response to external stimuli such as temperature⁷ or pH,⁸ specific interactions with biological systems,⁹ and in vivo degradability.¹⁰ Moreover, the coupling of peptides to other molecules, both natural and synthetic, expands the capabilities of both components in a synergistic manner. For example, limitations such as peptide solubility in aqueous systems can be overcome by conjugation to water soluble polymers.¹¹

Peptide conjugations are generally accomplished by grafting-to, a convergent method of peptide-conjugate synthesis in which both the peptide and target conjugate groups are synthesized individually and coupled there-after utilizing complimentary reactive sites on the conjugate and peptide, respectively.¹² Thiol-Michael type addition is a commonly utilized route towards peptide-conjugates because the highly reactive thiolate ion enables facile reactions with electron-deficient C=C bonds (e.g. maleimides, vinyl sulfones, acrylates & acrylamides) under mild conditions.¹³ Modification of peptide segments at the N-terminus with an acryloyl groups is an efficient method to synthesize peptide-conjugates via thiol-Michael type addition.¹⁴ In addition to peptide-conjugates, branched-peptide polymer structures are obtainable using reversible deactivation radical polymerization strategies to graft-through, or polymerize, the C=C bond of the acryloyl moiety.¹⁵

To approach applications that require lower cost materials for higher volume applications such as metal-recovery, membranes, and adhesives, there is a need for robust, versatile, and sustainable peptide synthetic methods. Solid or Liquid Phase Peptide Synthesis (SPPS/LPPS) is the current go-to strategy for synthesizing peptides 2–30 residues in length. With SPPS, amino acids are coupled sequentially, starting from an amino acid anchored to a polymer support.¹⁶ The same sequential coupling of amino acid residues is also accomplished by LPPS, which does not use a solid support. Protecting groups, such as tert-butyloxycarbonyl (BOC), are utilized to mitigate side reactions caused by nucleophilic amino functional groups.¹⁷ Since the invention of SPPS by Merrifield in 1963, significant improvements in protection chemistry, coupling reagents, and solvent choice has made SPPS/LPPS more efficient providing peptides in higher yields and purity.¹⁸ Nevertheless, the fundamental cycle of coupling, deprotection, and purification that comprises SPPS/LPPS requires excess reagents to achieve high coupling conversions and excess environmentally harmful solvents, like dimethylformamide, dichloromethane, and trifluoracetic acid, to complete the aforementioned steps.¹⁹

PCPS provides a green alternative to SPPS and LPPS. It is applicable in cases where peptides need not have uniform sequence and chain lengths to achieve a desired property. Since PCPS often results in average chain lengths (DP_n-avg) of about 5–10 residues, the conjugation to synthetic or natural polymeric chains can lead to synergistic interactions between the tethered chains that provides properties that mimic longer chain analogs.²⁰ Kinetically-controlled PCPS by cystine proteases (papain) is normally performed in aqueous buffer, using amino acid ethyl ester (AA-OEt) monomers, without protection-deprotection of N-terminus of the growing chain and amino acid (AA) functional R-groups (e.g. lysine, ε-amine).²¹ Generally, PCPS is conducted at a pH about 8 and 40 ^oC for seconds to hours with peptide yields between 30 and 80%.^22–24 Homo-oligopeptides,²⁵ co-oligopeptides with random,^26,27 block,²⁸ and alternating sequences,^29,30 terminal-modified oligopeptides,^31,32 and peptide-polymer conjugates have been prepared using PCPS.^33,34 Given published work thus far, PCPS is a promising method for scale-up in non-specialized reactors to provide peptides at costs compatible with higher volume specialty applications such as coating and specialty adhesives.³⁵

Modification of amino acids and peptide segments at the N-terminus with acryloyl moieties is an efficient method to synthesize peptide-conjugates with unique properties via grafting to or grafting through strategies.³⁶ However, peptides have traditionally been prepared by SPPS, which limits their use to highly specialized applications such as therapeutics.³⁷ Alternatively, kinetically-controlled PCPS can facilitate the synthesis of N-terminally functionalized oligopeptides in an aqueous, in a one-pot scheme. The polymerization of AA-OEt via PCPS relies on the reaction kinetics favoring aminolysis over hydrolysis.³⁸ As illustrated in Fig. 1, for aminolysis to be favored, the acryloyl grafter, N-acryl-(AA)_n-OEt (P₁ R-group in Fig 1), and the NH₂-(AA)_n-OEt acyl acceptor (P₂ R-group in Fig 1), must fit well within the S₁ and S₁’ catalytic site sub pockets, respectively, such that the acyl-enzyme complex (rate-limiting step) is formed.³⁹ Subsequent amide bond formation by reaction of acryloyl grafter (acyl donor) with NH₂-(AA)_n-OEt (n ≥ 1), acyl acceptor (aminolysis), must be kinetically preferred over hydrolysis for chains to propagate.

Figure 1.

Graphical representation of the mechanism for papain (cystine protease) catalyzed amide bond formation with acryloyl-grafter and AA-OEt substrates. Adapted with permission from Tsuchiya, K.; Numata, K. Chemoenzymatic Synthesis of Polypeptides for Use as Functional and Structural Materials. Macromol. Biosci. 2017, 17 (11), 1–15. Copyright © 2017 John Wiley and Sons.

Of the many proteases available, papain is often employed in PCPS due to its known broad substrate recognition. As such, papain has been employed in previous reports of N-acryloyl-peptide synthesis via Kinetically-controlled PCPS. In 2020, Tsuchiya et al. evaluated N-acryloyl-Ala-OEt as a grafter for papain-catalyzed peptide synthesis.⁴¹ As the ratio of N-acryloyl-Ala-OEt to NH₂-Ala-OEt monomer increased from 0.1:1 to 1:1 (mmol/mmol), the percent of precipitated peptide product that possessed an N-acryloyl terminus increased from 18 to >99%, whilst the total product %-yield decreased from 31.5 to 15.1% (by mass). Matrix-assisted laser desorption/ionization-time of flight (MALDI-TOF) revealed that N-acryloyl-terminated oligomers were formed that range in length from 4 to 10 residues, in which DP=5 was the strongest signal. Tsuchyia et al. also evaluated other AA-OEt monomers (Gly-OEt, Glu-[OEt]₂, Tyr-OEt, and Lys[ε-Boc]-OMe) at the 1:1 acryloyl-Ala-OEt/monomer molar ratio. Of the four, Gly-OEt yielded no precipitate while Glu-(OEt)₂, Tyr-OEt, and Lys(ε-Boc)-OMe all yielded precipitates with >99% acryloyl-oligopeptide terminal functionalization and product yields of 13.8%, 7.9%, and 55.3%, respectively.⁴¹ The above shows the potential of using N-acryloyl-AA-OEt grafters for the preparation of acryloyl-terminated oligopeptides. However, low yields at the 1:1 feed ratio where high N-terminal oligopeptide acryloylation was achieved is a concern.

Towards the objective of both high terminal chain N-acryloyl functionalization and yields, this study builds on previous work by expanding the grafters, monomers, and feed ratios. That is, the grafters studied herein include N-acryloyl-Glu-(OEt)₂, N-acryloyl-Leu-OEt, N-acryloyl-Ala-OEt; monomers studied are Leu-OEt and Glu-(OEt)₂; and the feed ratios of grafter-to-monomer include 0.1:1, 0.2:1, 0.4:1, and 1:1 (mmol/mmol). This provided the degree of structural and feed ratio variation that allowed interrogation of the relative importance of grafter and monomer structures to achieve both high N-acryloyl-oligopeptide functionalization and yield. Characterizations provided quantification of conversion of N-acryloyl-grafter to N-acryloyl-oligopeptide (% grafter conversion), the ratio of N-acryloyl-oligopeptide to NH₂-oligopeptide in the product (mole-% of terminally functionalized N-acryloyl-oligopeptide), and overall oligopeptide yield. Furthermore, DP_avg for both N-acryloyl-terminated and NH₂-oligopeptide were determined by MALDI-TOF. This work reveals unexpected results as to the factors that control efficient formation of N-acryloyl-terminated oligopeptides prepared by PCPS.

EXPERIMENTAL SECTION

Materials

l-Alanine ethyl ester hydrochloride (l-Ala-OEt·HCl), l-leucine ethyl ester hydrochloride (l-Leu-OEt·HCl), l-glutamic acid diethyl ester hydrochloride (l-(Et)₂-Glu·HCl), and >99% trifluoroacetic acid (TFA) were purchased from Chem-Impex. Acryloyl-chloride, sodium phosphate monobasic monohydrate, sodium phosphate dibasic anhydrous, sodium hydroxide, sodium bicarbonate, potassium chloride, >99.0% α-cyano-4-hydroxycinnamic acid (CHCA), 99.8% deuterated dimethylsulfoxide (DMSO_d6), silica gel (>60 Å, 70–230 mesh, 63–200 μm), and silica thin-layer chromatography (TLC) plates with fluorescent indicator were purchased from Millipore Sigma. Chloroform, methanol, ethyl acetate, n-hexanes, methylene chloride, acetonitrile, dimethylsulfoxide (DMSO), triethylamine (TEA), and concentrated hydrochloric acid were purchased from Merck Millipore. Papain was purchased from Acros. All chemicals were used as received without further purification.

General Method for the Synthesis of N-acryloyl-Amino Acid Ethyl Ester Grafters

As a representative example, the synthesis of N-acryloyl-Glu-(OEt)₂ is described using an adapted literature protocol.⁴¹ With magnetic stirring under an argon atmosphere, L-(Et)₂-Glu·HCl (7 g, 29.2 mmol) was dissolved in 30 mL of dry chloroform (CHCl₃) in a 150 mL round bottom flask, followed by addition of 2.2 equivalents (9 mL, 64.3 mmol) of TEA. The reaction mixture was cooled in an ice bath for 30 min. Next, 1.2 equivalents of acryloyl-chloride (2.83 mL, 35 mmol) in 20 mL CHCl₃ was added drop wise to this solution over a period of 30 min. Maintaining an argon atmosphere and magnetic stirring, the reaction mixture was allowed to warm to room temperature and continued for 16 h. Subsequently, the reaction mixture was transferred to a separatory funnel with 20 mL CHCl₃ and washed with saturated NaHCO₃ solution (2 × 30 mL), followed by KCl brine solution (2 × 30 mL), and lastly distilled water (2 × 30 mL). The organic layer was dried over MgSO₄, filtered, and the filtrate was concentrated by rotary evaporation. The oily residue was further purified by column chromatography (silica gel 200 mesh size; eluent 1:1 n-hexane/ethyl acetate). Following solvent evaporation, the oily residue crystallized on keeping in the freezer and residual solvent was removed under vacuum to yield a pale-yellow solid. Yield 5.3 g (71%), ¹H NMR (600 MHz) chemical shifts (δ in ppm) in DMSO-d₆ 1.18 (t, 6H), 1.86 (m, 1H), 2.00 (m, 1H), 2.38 (m, 2H), 4.04 (q, 4H), 4.33 (br, 1H), 5.63 (dd,1H), 6.11 (dd, 1H), 6.28 (m, 1H), 8.47 (d, 1H). HRMS (ESI-MS) m/z: [M + H]⁺ Calcd for C₁₂H₂₀NO₅ 258.13; Found 258.13.

Synthesis of N-acryloyl-Leu-OEt was performed by the above-described method giving a light amber solid. Yield 4.9 g (65%), ¹H NMR (600 MHz) chemical shifts (δ in ppm) in DMSO-d₆ 0.84 (d, 3H), 0.89 (d, 3H), 1.17 (t, 3H), 1.49–1.62 (m, 3H), 4.08 (q, 2H), 4.33 (m, 1H), 5.63 (dd,1H), 6.11 (dd, 1H), 6.28 (m, 1H), 8.43 (d, 1H). HRMS (ESI-MS) m/z: [M + H]⁺ Calcd for C₁₁H₂₀NO₃ 214.13; Found 214.14.

Synthesis of N-acryloyl-Ala-OEt was performed as described above giving a pale-yellow solid. Yield 4.6g (59%) ¹H NMR (600 MHz) chemical shifts (δ in ppm) in DMSO-d6 1.18 (t, 3H), 1.29 (d, 3H), 4.08 (q, 2H), 4.31 (m, 1H), 5.62 (dd,1H), 6.12 (dd, 1H), 6.26 (m, 1H), 8.49 (d, 1H). HRMS (ESI-MS) m/z: [M + H]⁺ Calcd for C₈H₁₄NO₃ 172.09; Found 172.10.

General Procedure for Protease-Catalyzed N-acryloyl-oligo(Glu)_x Synthesis

Glu-(OEt)·HCl (150.0 mg, 0.625 mmol) was transferred to a glass tube and dissolved in 4 mL of 1 M sodium phosphate buffer (PBS) (pH=8) with a Teflon-coated magnetic stir bar. N-acryloyl-Glu-(OEt)₂ (159.0 mg, 0.625 mmol) was dissolved in 1 mL of methanol and transferred into the reaction mixture. The pH of the reaction mixture was adjusted to 8.0 by adding 6 M NaOH_(aq). The mixture was incubated at 40°C for 15 min, at which point 10.4 U/mL papain (0.52 U/mg, 20 mg/mL) was added with vigorous mixing (see below the activity assay used). The reaction was allowed to stir for 1 h at 40°C. After 1 h, the precipitate was isolated via centrifugation at 10k rpm for 10 minutes. The supernatant was collected, mixed with equal volume of 0.1M HCl_(aq) to deactivate papain, and analyzed by MALDI-TOF. The white precipitate was washed once with 5 mL of 0.1M HCl_(aq) and once with 5 mL distilled water. After washing, the precipitate was lyophilized overnight, yielding white powder. The %-yield (by mass) of the recovered product was determined by dividing the total mass of substrate (N-acryloyl grafter and NH₂-AA-OEt) by the mass of the lyophilized product.

The monomer concentration is 0.125 M for all reactions. Grafter concentrations were adjusted according to the grafter to monomer molar ratios 0.1:1–1:1 (mmol/mmol). All reactions were run in triplicate and values reported are mean and standard deviation.

Protease Activity by Azocasein Assay

The method used to determine papain activity was adapted from a literature procedure.³⁴ In brief, to a 2 mL microcentrifuge tube (n=3), 0.25 mL of a 2.5% (w/v) azocasein solution (50 mM PBS, pH 8) was added. An additional 0.25 mL of 50 mM PBS (pH 8) was added to the blank. The microcentrifuge tubes were then placed in a 40 °C thermomixer to equilibrate for 15 mins. Enzyme stock solutions were prepared in 50 mM PBS (pH 8) at 1 mg/mL and, immediately thereafter, 0.25 mL of each enzyme stock solution was added to the appropriate microcentrifuge tubes with azocasein. After incubating for 30 min, 1 mL of 5.0% (v/v) trichloroacetic acid was added to each microcentrifuge tube, terminating the reaction via precipitation of nonhydrolyzed azocasein and protease. Each microcentrifuge tube was then centrifuged at 10k rpm for 10 min, and the supernatant was recovered and diluted with 1.5 mL of 0.5 M NaOH(aq). The absorbance of each solution was measured at 440 nm. Further dilution using 0.5 M NaOH(aq) was conducted when necessary to reach the linear region of the calibration curve. A calibration curve was created using dilutions of azocasein in 0.1 M NaOH(aq). The activity is reported in units per mg protease (U/ mg), where one unit is defined as the mass (mg) of azocasein hydrolyzed per minute. The activity of papain was determined to be 0.52 U/mg, respectively. All absorption measurements were acquired using disposable polystyrene cuvettes in a Jasco V-530 UV/vis at room temperature. Scans were made from 300 to 500 nm, with a medium response, a data pitch of 2 nm, and a scanning rate of 400 nm/min.

Nuclear Magnetic Resonance (NMR)

All ¹H NMR experiments for structural characterization of N-acryloyl-Glu-(OEt)₂, N-acryloyl-Leu-OEt, and N-acryloyl-Ala-OEt were completed on a Bruker Avance III 600 MHz standard-bore NMR with a ¹H cryoprobe and z-axis gradients in DMSO-d₆ (10 mg/mL). All ¹H NMR experiments for structural characterization of N-acryloyl-oligopeptide/NH₂-oligopeptide co-precipitate product were completed on the same NMR instrument using DMSO-d₆ with 5% trifluoroacetic acid (TFA) (5 mg/mL). All data was processed using Bruker Topspin 4.0.7. Proton chemical shifts were calibrated to DMSO_d6 at 2.50 ppm.

Matrix Assisted Laser Desorption/Ionization Time-of-Flight (MALDI-TOF)

MALDI-TOF was conducted on a Bruker Ultraflex III MALDI-TOF/TOF in positive-ion mode. Data acquisition was achieved using Bruker FlexControl software. Bruker Peptide Calibration Standard II was used to calibrate the instrument. A matrix solution was prepared using 10 mg/mL of α-cyano-4-hydroxycinnamic acid (CHCA) in ACN/water (1:1) with 0.1% (v) TFA. Samples were prepared by dissolving 1 mg of peptide in 200 μL of DMSO with 0.1% (v) TFA. A 10 μL aliquot was diluted in 240 μL of acetonitrile/water (1:1) with 0.1% (v) TFA. Supernatant samples were prepared by mixing 20 μL of supernatant with 230 μL of acetonitrile/water (1:1). Samples were spotted onto a stainless-steel target plate with 1 μL of acetonitrile/water peptide solution and 1 μL of CHCA matrix and dried under a stream of air. Acquired data was exported as mzXML files and further processed using open-source mass spectrometry software mMass (www.mmass.org) and Microsoft Excel. After baseline correction and smoothing, peaks were identified and labelled using theoretical masses of possible peptide products and adduct combinations. Peak data was used to determine number average (DP_n-avg) and weight average (DP_w-avg) degree of polymerization, as well as mean dispersity (Ð) values for all grafter/monomer combinations and NH₂-oligopeptides. The standard deviation for all dispersity values was ± 0.01.

Computational Modeling

Computational modeling was performed using PyRosetta, a Python-based interface to the Rosetta software suite.^44,45 Models were scored with the ref2015 scoring function, with added weights for constraints to favor models with geometries that correspond to effective active sites. Chemical structures for every amino acid ester substrate were created with Avogadro⁴⁶ and 200 conformers for each structure were generated using the conformational generation tool in Mercury by Cambridge Crystallographic Data Centre (CCDC), which references structures from the Cambridge Structural Database (CSD).⁴⁷

The papain-bound structures were generated as previously described.⁴⁸ Briefly, to generate starting structures, papain structure (PDB accession code 9pap)⁴⁹ was subjected to the Rosetta FastRelax protocol for initial minimization, while the acyl-enzyme structure from 2bu3⁵⁰ was used to generate constraints based on the geometry of the acyl intermediate, the nucleophilic acyl-acceptor, and the catalytic residues of papain. The acyl-acceptor and acyl-donor for each pair were initially docked into the active site based on the geometry of the acyl-enzyme structure using our previous model.⁴⁸

Models were generated via sampling of protein sidechain rotamers and ligand conformers to find compatible structures of the enzyme and substrates. Conformers for each substrate were sampled from the set of 200 generated for each structure. Constraints were applied to ensure the sampled conformers still adhered to the necessary geometry for the nucleophilic attack in the active site. The sampling consisted of rounds of repack-minimization of enzyme-substrate complexes to find sets of energetically favorable conformations of the active site subject to maintenance of catalytically viable geometry enforced via constraints. Sampling was repeated 20 times for each pair of acyl-donors and acyl-acceptors, which consists of five acyl-donors (N-acryloyl-Glu-(OEt)₂, N-acryloyl-Leu-OEt, N-acryloyl-Ala-OEt; NH₂-Glu-(OEt)₂, NH₂-Leu-OEt) and two acyl-acceptors (Leu-OEt and Glu-(OEt)₂). The conformations with the lowest energies for each complex were used for comparisons.

Analysis was conducted by comparing per-residue scores between models. Shell energies were calculated using a metric that summed up the energies of the residues in the active site pocket defined as all residues with C-α atoms < 8 Å from the acyl-acceptor and acyl-donor. Geometric constraints were applied to every case when generating the models, but not when scoring shell energies of the generated models.

All input files used for Rosetta calculations and computed interaction energy tables are provided as supplementary material.

RESULTS AND DISCUSSION

Generalized mechanism for papain-catalyzed synthesis of N-acryloyl-terminated oligopeptides

Figure 1 describes key reactions and intermediate of kinetically controlled PCPS.⁴⁰ Briefly, an acyl-enzyme complex is formed between the carbonyl of an amino acid ethyl ester (AA-OEt) and the thiolate of the cysteine within the protease active site. Then, amidation of the activated thiol ester occurs via a nucleophilic attack by a NH₂ moiety (e.g. α-amine) of an AA-OEt monomer or the corresponding elongated oligomer. To synthesize N-terminal modified oligopeptides via PCPS, the grafters used herein are N-acryloyl modified AA-OEt. Grafter conversion for functionalizing the N-terminus of oligopeptides depends on that it forms acyl-activated intermediates that react with the NH₂ moieties of a monomer (NH₂-AA-OEt) or oligomer (NH₂-[AA]_x-OEt) at rates that are competitive with that of corresponding monomer and oligopeptide acyl donors. Furthermore, the relative rate of amidation of acyl-enzyme intermediate must be high relative to hydrolysis. In a report published by Li et al., papain was found to be an exceptional catalyst for converting Glu-(OEt)₂ to oligo(γ-OEt-(Glu)_x-OEt_, providing the product in good yield (81%) within 15 minutes.²² Other reports for Leu-OEt^23,51 and Ala-OEt²⁵ support that papain-catalyzed PCPS can occur forming oligopeptides where amidation outcompetes hydrolysis reactions.

Grafter synthesis

The grafters studied herein were synthesized by modification of the N-terminus of the chosen amino acid esters with acryloyl-chloride to yield, N-acryloyl-Ala-OEt, N-acryloyl-Glu-(OEt)₂, and N-acryloyl-Leu-OEt. The ¹H NMR of each grafter, displayed in Figs. S1–S3, shows that the NMR signals and corresponding integration values are consistent with that of the expected products. In addition, molecular mass peaks (M+1) determined by ESI-MS are in agreement with the expected values. All three grafters were utilized in papain-catalyzed co-oligomerizations with the following monomers: Glu-(OEt)₂ and Leu-OEt.

Rational for PCPS-catalyzed oligomerization reaction conditions

Reaction conditions (1 M PBS, 20% (v/v) MeOH, pH=8, 40 °C) were based on previously published literature.^41,42 The chosen temperature for the oligomerizations described herein is based on a previous study where variation of temperature was evaluated for the synthesis of oligo(Phe)_x using papain catalyst.⁴³ To obtain homogeneous reaction conditions at the start of co-oligomerizations for monomers Glu-(OEt)₂ and Leu-OEt, the total concentration of grafter and monomer did not exceed 0.25 M, a favorable concentration established in previous work for PCPS of a series of hydrophobic AA-OEt monomers.²³ The monomer concentrations for Leu-OEt and Glu-(OEt)₂ were capped at 0.125 M. Grafter concentrations in experiments with Glu-(OEt)₂ and Leu-OEt monomers were changed to values that gave grafter to monomer feed ratios of 0.1:1, 0.2:1, 0.4:1, and 1:1 (mmol/mmol). The rational for not decreasing the grafter to monomer ratio below 0.1:1 mmol/mmol is that, generally, the maximum DP_n-avg of PCPS catalyzed oligopeptides is about 10 residues.⁴⁰

Furthermore, the aqueous media was modified with 20% (v/v) methanol co-solvent, so grafter solubility was maintained at high ratios of grafter-to-monomer (i.e., 1:1 mmol/mmol). Previous work by our group revealed that the addition of 20% methanol co-solvent to papain-catalyzed synthesis of NH₂-oligo(Phe)_y resulted in an increase in %-yield from 30 to 50%.²³ For all three grafters, reactions conducted at the 1:1 (mmol/mmol) feed ratio were translucent suspensions prior to the addition of the protease. Decreasing the concentration of N-acryloyl grafter below 1:1 (mmol/mmol) provided homogeneous solutions. In all experiments, the addition of papain caused the precipitation of a white solid in < 1 min. After 1 h, the insoluble fraction (co-precipitate) was isolated by centrifugation, washed, and lyophilized to yield a white powder consisting of both N-acryloyl-oligopeptides and NH₂-oligopeptides. Initial confirmation of amidation via papain-catalysis was supported by an observed decrease in reaction pH from 8.0 to 7.4 for all reactions due to liberation of the HCl salt from the AA-OEt monomer. The average %-yield (by mass) of all reactions and the average %-yield of N-acryloyl oligopeptide are reported in Fig 3 (C and D). These values will be discussed later.

Figure 3.

Values for (A) mol-% grafter conversion, (B) %-yield (by mass) N-acryloyl-oligopeptide, (C) mol-% N-acryloyl-oligopeptide in the insoluble fraction, and (D) %-yield (by mass) of the insoluble fraction of all grafter-monomer combinations and feed ratios. N-acryloyl-Ala-OEt, N-acryloyl-Glu-(OEt)₂, and N-acryloyl-Leu-OEt are identified by blue triangles, green squares, and yellow circles, respectively. The monomers Glu-(OEt)₂ and Leu-OEt are indicated by dashed lines colored purple, and orange, respectively. In some cases, the value of standard deviation is less than the size of the symbol denoting the mean value. The axis in panel C was re-scaled for clarity.

Analysis of chain length

The isolated insoluble fraction (precipitate) of each N-acryloyl grafter-monomer combination was first analyzed using MALDI-TOF to i), identify the presence of N-acryloyl oligopeptides and ii), determine the distribution of chain lengths of N-acryloyl-oligopeptides/NH₂-oligopeptides. Analysis of the soluble fraction via MALDI-TOF did not reveal any soluble N-acryloyl/NH₂-oligopeptides. Using the MALDI-TOF peak intensities of identified signals, the number average degree of polymerization (DP_n-avg), weight average degree of polymerization (DP_w-avg), and dispersity (Ð) of both N-acryloyl-oligopeptide and NH₂-oligopeptides in the isolated co-precipitate were determined. The results are summarized in Table 1 and Table S2. The DP_avg refers to the number of amino acid monomers directly added to the N-acryloyl grafter (N-acryloyl-oligopeptides) and the number of amino acid monomer units in the chain (NH2-olipeptides). The DP_avg values determined by MALDI-TOF of N-acryloyl grafter and monomer combinations are displayed in Table 1. Since determined DP_avg values of N-acryloyl and NH₂-terminated oligopeptides show low variation across the different feed ratios, Table 1 lists mean and standard deviation of DP_avg and Đ values calculated from those at the different feed ratios studied herein. The low deviation in DP_avg and Đ values across feed ratios is supported by the low standard deviation values.

Table 1.

DP_avg values of grafted (AA)_x-OEt segments of N-acryloyl-oligopeptides for all grafter/monomer combinations and feed ratios determined using MALDI-TOF. The cumulative DP_avg and dispersity (Đ) values are the average from experiments at all feed ratios. Reactions were conducted in 1 M phosphate buffer with 20% MeOH, pH=8, 40 °C, for 1 h, using papain (10.5 U/mL, 20 mg/mL) as catalyst.

Grafter	Monomer	DPavg N-acryloyl-Oligopeptide (MALDI-TOF)
		0.1 : 1	0.2 : 1	0.4 : 1	1 : 1	Cumulative DPavg	Cumulative Đavg
N-acryloyl-Ala-OEt	Glu-(OEt)2	7.0 ± 0.3	7.0 ± 0.2	6.7 ± 0.3	6.2 ± 0.3	6.7 ± 0.4	1.04
	Leu-OEt	4.7 ± 0.1	4.8 ± 0.1	4.4 ± 0.1	4.5 ± 0.1	4.6 ± 0.2	1.02
N-acryloyl-Glu-(OEt)2	Glu-(OEt)2	6.4 ± 0.1	6.2 ± 0.1	5.8 ± 0.1	6.3 ± 0.1	6.2 ± 0.3	1.03
	Leu-OEt	4.0 ± 0.1	4.0 ± 0.1	3.9 ± 0.1	4.1 ± 0.1	4.1 ± 0.2	1.05
N-acryloyl-Leu-OEt	Glu-(OEt)2	6.1 ± 0.6	6.2 ± 0.4	6.2 ± 0.5	7.0 ± 0.1	6.3 ± 0.4	1.03
	Leu-OEt	4.0 ± 0.1	3.9 ± 0.2	3.8 ± 0.3	3.6 ± 0.4	3.8 ± 0.1	1.02

Labelled MALDI-TOF spectra are displayed in Figure 2. Regardless of the feed ratio for the monomer Glu-(OEt)₂ and grafters studied herein (N-acryloyl-Ala-OEt, N-acryloyl-Glu-(OEt)_2, and N-acryloyl-Leu-OEt) a series of signals were observed in the MALDI-TOF spectra that correlate to N-acryloyl/NH₂-oligo(Glu)_y species (Figure 2 A–C) with different adducts (Na⁺ and K⁺) and C-terminus (-CO₂Et or –CO₂H). Peaks were observed from 900–2000 m/z separated by 157 m/z, which correlates to the mass of a glutamic acid residue.²² Observation of the MALDI-TOF reveals that 5–10 Glu residues were directly added to the C-terminus of each grafter via papain-catalysis, resulting in a DP_n-avg of 6.7 ± 0.4 for N-acryloyl-Ala-OEt, 6.2 ± 0.3 for N-acryloyl-Glu-(OEt)₂, and 6.3 ± 0.4 for N-acryloyl-Leu-OEt as shown in Table 1. These data indicate that precipitation of the kinetically favored products results once the oligopeptide reaches a critical DP.Discussion of the effect of the N-acryloyl group on DP_avg of the N-acryloyl-oligopeptide in comparison to NH₂-oligopeptide is provided at the end of this section.

Figure 2.

MALDI-TOF spectra of the insoluble fraction for each combination of N-acryloyl/NH₂-oligopeptides synthesized via papain catalysis (10.4 U/mL) in 1 M phosphate buffer 20% methanol (pH 8) at 40 °C for 1 h at feed ratio 1:1 grafter to monomer: N-acryloyl-Ala-OEt/Glu-(OEt)₂ (A) and Leu-OEt (D), N-acryloyl-Glu-(OEt)₂/Glu-(OEt)₂ (B) and Leu-OEt (E), N-acryloyl-Leu-OEt/Glu-(OEt)₂ (C) and Leu-OEt (F). Circles denote the N-acryloyl-oligopeptide species (Na⁺, -CO₂Et at theC-terminus), whereas squares represent the major NH₂-oligopeptide species (Na⁺, -CO₂Et at theC-terminus). Integer values represent the DP.

As shown in Table 1, exchanging the monomer from Glu-(OEt)₂ to Leu-OEt resulted in shorter N-acryloyl-oligopeptides. This is explained by the greater self-assembling potential of Leu residues which results in oligopeptide precipitation at shorter chain lengths.²³ Peaks from 500–1000 m/z, separated by 113 m/z (mass of Leu residue), were observed in all MADLI_TOF spectra corresponding to N-acryloyl/NH₂-oligo(Leu)_y peptides (Figure 2 D–F). A range of 3–7 Leu residues were added to N-acryloyl-Ala-OEt, N-acryloyl-Glu-(OEt)₂, and N-acryloyl-Leu-OEt. The corresponding MALDI-TOF determined DP_avg values are 4.6 ± 0.2, 4.1 ± 0.2, and 3.8 ± 0.1, respectively.

The chain lengths of NH₂-oligopeptide co-product was also determined using MALDI-TOF and DP_avg values are provided in Table S2. Based on MALDI-TOF, the DP_avg of NH₂-oligo(Glu)_y, co-synthesized in the presence of N-acryloyl-Ala-OEt, N-acryloyl-Glu-(OEt)₂, and N-acryloyl-Leu-OEt grafters, were 8.4 ± 0.2, 8.2 ± 0.5, and 8.3 ± 0.4, respectively. In contrast, the DP_avg based on MALDI-TOF of NH₂-oligo(Leu)_y, co-synthesized in the presence of N-acryloyl-Ala-OEt, N-acryloyl-Glu-(OEt)₂, and N-acryloyl-Leu-OEt grafters, were 6.4 ± 0.1, 6.8 ± 0.2, and 5.9 ± 0.1, respectively. For control experiments where NH₂-oligo(Glu)_y was synthesized in the absence of grafter but in the same experimental conditions, the DP_avg of NH₂-oligo(Glu)_y, determined by MALDI-TOF were 8.5 ± 0.2. The DP_avg of NH₂-oligo(Glu)_y correlates with those above when the oligopeptide was synthesized in the presence of grafters as well as previous studies on papain-catalyzed NH₂-oligo(Glu)_y synthesis.^22,23 For NH₂-oligo(Leu)_y, the DP_avg for control experiments, determined by MALDI-TOF is 6.6 ± 0.3. The DP_avg values of NH₂-oligo(Leu)_y synthesized both in the presence and absence of grafter also correlate well to other reports of NH₂-oligo(Leu)_y synthesis via papain-catalysis.^23,51 MALDI-TOF spectra of both NH₂-oligo(Glu)_y (Figure S16) and NH₂-oligo(Leu)_y (Figure S18) are provided in the Supporting Information.

Precipitation of the N-acryloyl-oligopeptide/NH₂-oligopeptide occurs via aggregation or coassembly of different oligopeptide chains of a certain length, defined as the DP_avg. Comparison of the DP_avg between reaction products when the N-acryloyl-amino acid is included reveals that the DP_avg of N-acryloyl-oligopeptides is 1 less amino acid in length than the NH₂-oligopeptide product (except for N-acryloyl-Glu-OEt:Leu-OEt, where the difference in DP_avg between N-acryloyl-Glu-(Leu)_x/NH₂-oligo(Leu)_y is 2). For example, the DP_avg of N-acryloyl-Glu-(Glu)_x, including the N-acryloyl-Glu residue, is about 7, whereas the DP_avg of NH₂-oligo(Glu)_y is 8 residues. We hypothesize the N-acryloyl moiety contributes hydrophobicity like that of an amino acid residue, thus limiting the DP_avg of N-acryloyl-oligopeptides equal to the DP_avg of the NH₂-oligopeptide minus one. Elucidation of the mechanism of coassembly/aggregation between N-acryloyl-oligopeptides/NH₂-oligopeptides was not carried out.

Analysis of Grafter Efficiency

Herein, the grafter efficiency is defined by how much grafter is converted to N-acryloyl-oligopeptide. Other values, including the mol-% of N-acryloyl-oligopeptide in the co-precipitate and %-yield (by mass) of the N-acryloyl-oligopeptide, provide additional insights into grafter efficiency. Using ¹H NMR peak integration in combination with MALDI-TOF data, the mole ratio of N-acryloyl-oligopeptides to NH₂-oligopeptides (Eq. S1), NH₂-oligopeptide DP_avg (Eq. S6), mass fraction of N-acryloyl-oligopeptide (Eq. S9), grafter conversion (mol-%) (Eq. S13) were calculated. Values of %-yield N-acryloyl-oligopeptide, N-acryloyl-oligopeptide content in the insoluble fraction (mol-%), and grafter conversion (mol-%) determined at varying feed ratios of grafter to monomer are displayed in Figure 3.

Labelled ¹H NMR spectra are provided in the Supporting Information (Figures S5–10). In all grafter-monomer combinations, signals corresponding to N-acryloyl protons are observed at 5.6 ppm, 6.1 ppm, and 6.3 ppm. Overlapping signals corresponding to α-methine protons of both N-acryloyl and NH₂-oligopeptides are observed around 4.1–4.3 ppm. The N-terminal α-methine protons of NH₂-oligo(Glu)_y and NH₂-oligo(Leu)_y are observed at 3.8 ppm and 4.4 ppm, respectively. Other important signals identified in ¹H NMR spectra are the β-methylene protons of Leu at 1.4 ppm and the γ-methylene protons of Glu at 2.3 ppm.

The mol-% of N-acryloyl-oligopeptide (Figure 3B) for all grafter-monomer combinations was determined by dividing the average integration of all N-acryloyl proton signals by the sum of both the average N-acryloyl proton integration value and NH₂-oligopeptide N-terminus α-methine proton integration value (Eq S1). For all grafter-monomer combinations studied herein, the mol% of N-acryloyl-oligopeptides was significantly less than 99% observed previously by Tsuchiya and Numata using the grafter N-acryloyl-Ala-OEt with monomers Ala-OEt, Glu-(OEt)₂, Lys(Boc)-OMe, and Tyr-OET at 1:1 grafter to monomer feed ratio.²⁸ Observation of Figure 3B shows that, as the ratio of grafter-to-monomer was increased from 0.1 to 1, the mol-% N-acryloyl-oligopeptide in the insoluble fraction also regularly increased. Papain-catalyzed synthesis using Glu-(OEt)₂ with N-acryloyl-Leu-OEt and N-acryloyl-Ala-OEt grafters resulted in similar mol-% of N-acryloyl-oligopeptide. When N-acryloyl-Glu-(OEt)₂ was paired with Glu-(OEt)₂ monomer at 0.1:1 feed ratio, 21 ± 1 % of the insoluble fraction was N-acryloyl-oligo(Glu)_x and this value increased to 67 ± 3 % at 1:1 grafter to monomer. A similar trend was observed when the monomer was exchanged for Leu-OEt. However, the mol%-N-acryloyl-Glu-oligo(Leu)_x increased from 44 ± 1% to 81 ± 6 % at 0.1:1 and 1:1 feed ratio, respectively. Hence, N-acryloyl-Glu(OEt)-(Leu)_x formation occurs at a greater frequency than N-acryloyl-Glu(OEt)-(Glu-OEt)_y. This result was unexpected given that both Glu-(OEt)₂ and Leu-OEt have similar reactivity for Papain.²⁶

Determination of the mol% N-acryloyl-oligopeptide in the insoluble fraction by ¹H NMR and determination of the DP_avg of the N-acryloyl-oligopeptide by MALDI-TOF enables calculation of the amount of grafter (mol-%) converted to N-acryloyl-oligopeptide (Eq. S13). Values of grafter conversion are displayed in Figure 3A. Observation of Figure 3A shows that, conversion for all three grafters with Glu-(OEt)₂ monomer was consistent and did not exceed 15 % across all feed ratios. In contrast, at 0.1:1 grafter/monomer, with Leu-OEt as monomer, grafter conversions with N-acryloyl-Glu-(OEt)₂, N-acryloyl-Ala-OEt and N-acryloyl-Leu-OEt were 82 ± 2.7 %, 64 ± 1.3 % and 54 ± 2.6 %, respectively. Increasing the grafter-to-monomer ratio from 0.1:1 to 1:1 resulted in decreasing, yet similar values of grafter conversion amongst the three grafters with Leu-OEt monomer, such that at the 1:1 feed ratio 7.7 ± 0.8 %, 10 ± 1.7 %, and 15 ± 0.2 % of N-acryloyl-Glu-(OEt)₂, N-acryloyl-Ala-OEt, and N-acryloyl-Leu-OEt was converted, respectively. Hence, NH₂-Leu-OEt or NH₂-(Leu)_y-OEt is a better acyl-acceptor than NH₂-Glu-(OEt)₂ or NH₂-(Glu-OEt)_y-OEt with the N-acryloyl grafters studied herein. Moreover, these data indicates that grafter conversion is dependent on the monomer (acyl-acceptor) more so than the grafter (acyl-donor). Computational modeling which examines energetic interactions between grafters and monomers within the papain active site was performed and the results are discussed in the section below.

The mass fraction of N-acryloyl/NH₂-oligopeptide in the insoluble fraction was calculated using Eq. S9 and Eq. S10. The %-yield of N-acryloyl-oligopeptide (Figure 3C) was calculated by the product of Eq. S9 with the mass of the insoluble fraction. The %-yield of N-acryloyl-oligopeptides when Glu-(OEt)₂ was used as monomer resulted in minimal yield, < 15 % for all grafters. The monomer Leu-OEt resulted in improved %-yield of N-acryloyl-oligopeptides, yet the maximum %-yield of N-acryloyl-oligopeptide did not exceed 30 %. In general, as the concentration of grafter increased in the feed with Leu-OEt monomer, the %-yield of N-acryloyl-oligopeptide decreased.

Control experiments were conducted to probe the effect of N-acryloyl grafters on papain catalytic activity. First, NH₂-oligo(Glu)_y and NH₂-oligo(Leu)_y were synthesized under the same conditions studied herein using 0.125 M substrate in the absence of grafter. Catalytic activity was considered by comparison of the %-yield of NH₂-oligo(Glu)_y and NH₂-oligo(Leu)_y synthesized with and without grafter present. The %-yield of NH₂-oligo(Glu)_y and NH₂-oligo(Leu)_y synthesized without grafter present was 47 ± 0.2 % and 82 ± 0.6 %, respectively. In comparison, at a feed ratio 0.1:1 (in which the concentration of N-acryloyl grafter is minimum), the total %-yield of papain-catalyzed grafter/monomer reactions using Glu-(OEt)₂ monomer does not exceed 30% (Figure 3D). When Leu-OEt is chosen as the monomer, %-yield does not exceed 62% except when N-acryloyl-Ala-OEt is the grafter (Figure 3D). This outlier is most likely due to preferable active site/subsite pocket binding for Leu residues within papain, more so than Ala residues, thus favoring NH₂-oligo(Leu)_y formation. Also, the possibility that N-acryloyl groups react to some extent via thiol-Michael type addition, thus reducing papain activity, was considered.^52,53 The N-acryloyl moiety was substituted by the N-propionyl saturated analogue by synthesizing N-propionyl-Glu-(OEt)₂. Papain-catalyzed grafter/monomer reactions involving N-propionyl-Glu-(OEt)₂ were limited to the greatest feed ratio (1:1). The N-propionyl-Glu-(OEt)₂ grafter obtained post-synthesis was an oil, which when mixed with the reaction media, resulted in a heterogenous mixture. The heterogeneity of the reaction at high grafter to monomer ratio (1:1) is expected based on solubility of the N-acryloyl grafters at the same feed ratios. Results in Table S2 reveal that the relative grafter conversion and total yield of co-precipitate using N-propionyl-Glu-(OEt)₂ and the monomer Glu-(OEt)₂ was 1.4 ± 0.4% and 4.5 ± 0.8%. When Leu-OEt was employed as the monomer with N-propionyl-Glu-(OEt)₂, 9.7 ± 0.3% of the grafter was converted to N-acryloyl-oligopeptide and the %-yield was 20.8 ± 0.7%. These values are similar to values obtained with N-acryloyl-Glu-(OEt)₂ grafter and Glu-(OEt)₂/Leu-OEt monomers at 1:1 feed ratio. Overall, the data discussed suggests the presence of N-acryloyl grafters does effect papain activity, however, it may not be through a thiol-Michael addition mechanism. Further experimentation is needed to establish how electron-deficient N-acryloyl grafters interact with thiols in protease active sites.

Rationalizing Observed Substrate Preferences Using Molecular Modeling

To investigate the dependency of grafter conversion on the identity of the acyl-acceptor monomer, computational modeling was performed with Rosetta modeling software as described in the methods section. Acyl-enzyme models were generated to determine energies of substrate binding of different acyl-donor:acyl-acceptor pairs to the papain active site. All acyl-donors and acyl-acceptors are represented as single monomers (e.g., NH₂-AA-OEt) for the purposes of modeling, but the acyl-acceptors may also exist as short oligomers (e.g., NH₂-[AA]_x-OEt).

Computed energies of the active site residues and first-shell interactions were used as proxies for activity. These energies are calculated based on summing up the energies of interactions between residues while maintaining a near-attack conformation for peptide bond formation. A near-attack geometry was enforced in the modeling by using rigid-body positional restraints for the acyl-enzyme intermediate and the acyl acceptor. A lower shell energy would represent a more stable pocket, which is expected to result in higher activity. In all three N-acryloyl grafter cases, the model with the Leu-OEt monomer has a lower shell energy than the model with the Glu-(OEt)₂ monomer, which is consistent with experimental data showing that using the Leu-OEt monomer results in greater grafter conversion (Table 2). The Leu-OEt monomer results in a 5.5 REU difference for the N-acryloyl-Ala-OEt grafter, a 4.9 REU difference for the N-acryloyl-Glu-(OEt)₂ grafter, and a 0.6 REU difference for the N-acryloyl-Leu-OEt grafter (Table 2). Thus, the developed modeling pipeline can generate conformations and energy values in qualitative agreement with experimental data.

Table 2.

Energies of the pockets in each N-acryloyl grafter model in REU (Rosetta Energy Units). Pockets were defined by the grafter, the monomer, the catalytic residues, and residues neighboring the grafter and monomer.

Grafter	Monomer	Pocket Energy (REU)
N-acryloyl-Ala-OEt	Glu-(OEt)2	−14.8
	Leu-OEt	−20.3
N-acryloyl-Glu-(OEt)2	Glu-(OEt)2	−9.5
	Leu-OEt	−14.4
N-acryloyl-Leu-OEt	Glu-(OEt)2	−21.3
	Leu-OEt	−21.9

Differential Interactions in various N-acryloyl-AA-OEt:AA-OEt Grafter Models

Having recapitulated the improved efficiency of grafter conversion with the Leu-OEt monomer over the Glu-(OEt)₂ monomer, energies of individual residues in the pocket were compared between models to gain further insight into the underlying interactions responsible for the observed preference for Leu-OEt (Table 3; Figure 4).

Table 3.

Energies of the top residues with the largest energy differences (in REU) between the Glu-(OEt)₂ and Leu-OEt monomer models for the N-acryloyl grafters.

Grafter	Residue	Energy Score (REU)
		With Glu-(OEt)2	With Leu-OEt	Difference
N-acryloyl-Ala-OEt	Gln-19	3.8	1.2	2.6
	Ala Grafter	3.8	2.0	1.8
	Gly-23	2.2	0.6	1.6
	Monomer	−0.9	−1.6	0.7
N-acryloyl-Glu-(OEt)2	Monomer	0.6	−1.3	1.9
	Asp-158	−2.0	−3.2	1.2
	Ala-136	−2.1	−2.9	0.8
	Ala-137	1.3	0.8	0.5
	Val-161	−2.5	−3.0	0.5
N-acryloyl-Leu-OEt	Gln-19	3.3	1.9	1.4
	Monomer	−1.1	−2.4	1.3
	Ala-136	−2.1	−2.7	0.6
	Trp-177	0.1	−0.4	0.5

Figure 4.

Active site docking models with the N-acryloyl-AA-OEt grafters. For the N-acryloyl-Ala-OEt grafter, the four residues with the largest differences in energy between the models with the Glu-(OEt)₂ monomer (A) and the Leu-OEt monomer (B) with the N-acryloyl-Ala-OEt grafter are Gln-19, Gly-23, the Ala-based grafter, and the monomer. Two residues that helped contribute to the largest pocket energy differences are Ala-136 and Cys-22. For the N-acryloyl-Glu-(OEt)₂ grafter, five residues with the largest differences in energy between the models with the Glu-(OEt)₂ monomer (C) and the Leu-OEt monomer (D) with the N-acryloyl-Glu-(OEt)₂ grafter are Ala-136, Ala-137, Asp-158, Val-161, and the monomer. One residue that helped contribute to the largest pocket energy differences is His-159, which has a chi 2 dihedral value of 67.3° with the Glu-(OEt)₂ monomer (C) and 63.5° with the Leu-OEt monomer (D). For the N-acryloyl-Leu-OEt grafter, the four residues with the largest differences in energy between the models with the Glu-(OEt)₂ monomer (E) and the Leu-OEt monomer (F) with the N-acryloyl-Leu-OEt grafter are Gln-19, Ala-136, Trp-177, and the monomer. Two residues that helped contribute to the largest pocket energy differences are Ser-176 and Ala-137.

For models with N-acryloyl-Ala-OEt grafters, the residues that are more energetically stable with the Leu-OEt monomer are Gln-19, the N-acryloyl-Ala-OEt grafter, Gly-23, and the nucleophilic monomer. The modeling produces a conformation of the enzyme-substrate complex that is constrained to be in the near-attack position, thus some residue energies are positive. However, comparative analyses of individual residue energies in the presence of Leu-OEt and Glu-OEt monomers is indicative of the relative stabilization/destabilization of the active conformation. For example, in the model with Leu-OEt monomer, Gln-19 can hydrogen-bond with both monomers through its amine moiety and the monomer α-ester carbonyl oxygen, but the conformation of the Glu-(OEt)₂ monomer causes Gln-19 to rotate to accommodate the hydrogen bond, thus rotating it out of the optimal geometry for the active site resulting in a higher energy value for the per-residue energy in the Glu-OEt model (Table 3, Figure 4A). The rotamer of Gln-19 adopted with Leu-OEt also allows for an additional hydrogen bond with Cys-22 (Figure 4B) that is absent with the rotamer adopted with Glu-(OEt)₂. Similarly, both monomers have negative energy values, but the Leu-OEt monomer is more stabilized. One such stabilizing interaction could be from the isobutyl group on the Leu-OEt monomer being more favorable for the hydrophobic Ala-36. The grafter is also less destabilized with Leu-OEt than with Glu-(OEt)₂. The ester and amine groups of Leu-OEt are closer to the carbonyl of the N-acryloyl-Ala-OEt grafter than those of Glu-(OEt)₂, with the hydrophobic isobutyl group on Leu-OEt tilted away from the amide group on the grafter. The positioning of the Leu-OEt monomer also causes the acryloyl group on the grafter to be further from Gly-23, thereby decreasing repulsion.

Similar analysis of models with N-acryloyl-Glu-(OEt)₂ grafters reveals that the residues that are more energetically stable with the Leu-OEt monomer are the monomer itself, Asp-158, Ala-136, Ala-137, and Val-161 (Table 3; Figure 4 C,D). The Leu-OEt monomer is stabilized via an additional a hydrogen bond between its amine hydrogen and the backbone carbonyl of Asp-158 (Figure 4D). Meanwhile, the carbonyl oxygen on the γ-ester on the Glu-(OEt)₂ monomer is pointing towards the β-carboxylic acid (Figure 4C). Both monomers form hydrogen bonds with the nitrogen on the ring of His-159, which results in very slightly different His-159 conformations (difference of 67.3° – 63.5° = 3.8° in chi 2) due to the different positions of the monomers (Figure 4 C,D). The slightly altered position of His-159 affects its repulsion with Val-161, as the repulsion score between the two residues decreases by 1.0 REU with the Leu-OEt monomer.

For models with N-acryloyl-Leu-OEt grafters, the difference in shell energies is not as pronounced as with the other grafters. However, the shell energy with the Leu-OEt monomer is still lower than with the Glu-(OEt)₂ monomer, especially for Gln-19, the nucleophilic monomer, Ala-136, and Trp-177. (Table 3; Figure 4 E,F). Overall, we observe both direct stabilization of substrates as well as indirect stabilization of the active site residues in models with Leu-OEt monomers compared to Glu-OEt monomers with all three grafters. Indirect effects are observed because interactions from the monomer can also cascade into affecting other residues indirectly. The distortion by the monomer of one residue can subsequently disrupt its interactions with its neighboring residues. All these interactions add up to affect stability of the active site.

Experiments with Leu-OEt and Glu-(OEt)₂ monomers without N-acryloyl grafters demonstrate no preference for either monomer, which differs from the observed preference for Leu-OEt with the N-acryloyl grafters. Because of this, models with N-acryloyl grafters were compared with models with NH₂-acyl donors to investigate possible reasons for this emergence of a preference of acyl-acceptor based on the type of acyl-donor. For each NH₂-acyl donor, both models, one of each acyl acceptor, have similar shell energies, which is also consistent with experimental data of equal preference of either Leu-OEt and Glu-(OEt)₂ (Table 4). The Leu-OEt acyl-acceptor results in a small 0.9 REU difference for the Glu-(OEt)₂ acyl-donor and a 1.2 REU difference for the Leu-(OEt) acyl-donor, which are not as large energy differences as for with the N-acryloyl-Ala and N-acryloyl-Glu grafters (Table 2, Table 4). Residue-wise breakdown of energy differences for these models is provided in the Supporting Information.

Table 4.

Energies of the pockets in each NH₂-acyl donor model in REU (Rosetta Energy Units). Pockets were defined by the acyl-donor, the acyl-acceptor, the catalytic residues, and residues neighboring the grafter and monomer.

Acyl-Donor	Acyl-Acceptor	Pocket Energy (in REU)
Glu-(OEt)2	Glu-(OEt)2	−6.6
	Leu-OEt	−7.5
Leu-OEt	Glu-(OEt)2	−3.8
	Leu-OEt	−5.0

Conclusion

This work revealed important factors controlling the efficiency of preparing N-acryloyl functionalized oligopeptides by kinetically controlled PCPS using papain as catalyst. Different combinations of three grafters (N-acryloyl-Ala-OEt, N-acryloyl-Glu-(OEt)₂, N-acryloyl-Leu-OEt) and two monomers (Leu-OEt and Glu-(OEt)₂) at grafter-to-monomer molar ratios of 0.1:1, 0.2:1, 0.4:1, and 1:1 (mmol/mmol) served as variables.

Characterizations of ¹H NMR and MALDI-TOF spectra enabled determination DP_avg of N-acryloyl-oligopeptides and NH₂-oligopeptides, N-acryloyl-grafter conversion to N-acryloyl-oligopeptide (% grafter conversion), the ratio of N-acryloyl-peptide to NH₂-oligopeptide in the product (mole-% N-acryloyl-oligopeptide), and overall N-acryloyl-oligopeptide yield. Litte variation in DP_avg between products was observed, likely because these values are mainly dependent on kinetically controlled precipitation of oligomers once the surpass critical chain lengths.

Conversion for all three grafters with Glu-(OEt)₂ monomer was consistent and did not exceed 15% across all feed ratios. In contrast, with Leu-OEt as monomer at 0.1:1 grafter/monomer feed ratio, grafter conversions with N-acryloyl-Glu-(OEt)₂, N-acryloyl-Ala-OEt and N-acryloyl-Leu-OEt were 82 ± 2.7 %, 64 ± 1.3 % and 54 ± 2.6 %, respectively. Consistent with mol-% values, the %-yield by mass of N-acryloyl-oligopeptides, synthesized using Glu-(OEt)₂ and the three grafters, ranged from 1% to 8% whereas, using Leu-OEt as monomer, the corresponding %-yield by mass values varied from 15% to 30%. Hence, NH₂-Leu-OEt is a more efficient acyl-acceptor than NH₂-Glu-(OEt)₂ for the series of N-acryloyl grafters studied herein. Grafter conversion of N-propionyl-Glu-(OEt)₂ with Glu-(OEt)₂ and Leu-OEt monomers (1:1 feed ratio) using papain catalyst was 1.4 ± 0.4% and 9.7 ± 0.3%, respectively. These data are similar to grafter conversion values obtained with N-acryloyl-Glu-(OEt)₂ and Glu-(OEt)₂/Leu-OEt monomers suggesting that N-acryloyl-Glu-(OEt)₂ does not react with the papain active site thiolate ion.

Computational modeling was able to qualitatively recapitulate the experimental findings that for the monomer:grafter combinations studied here, grafter conversion depends on the monomer rather than the grafter structure. Rosetta-based models showed pocket energy differences of 5.5 REU for N-acryloyl-Ala-OEt, 4.9 REU for N-acryloyl-Glu-(OEt)₂, and 0.6 REU for N-acryloyl-Leu-OEt, all in favor of the Leu-OEt monomer over Glu-(OEt)₂. Except for the N-acryloyl-Leu-OEt models, these differences are larger than those with the NH₂-acyl donor models (energy differences of 0.9 REU for NH₂-Glu-(OEt)₂ and 1.2 REU for NH₂-Leu-OEt). Thus, this work reveals that, for the monomer-grafter series studied herein, the grafter conversion is controlled by the monomer acyl-acceptor structure and placement in the active site, not the grafter acyl-donor structure.

List of Abbreviations

SPPS

Solid-Phase Peptide Synthesis

PCPS

Protease-Catalyzed Peptide Synthesis

LPPS

Liquid-Phase Peptide Synthesis

BOC

tert-butyloxycarbonyl

DP_n-avg

Number Average Degree of Polymerization

DP_w-avg

Weight Average Degree of Polymerization

Ð

Dispersity

DP

Degree of Polymerization

AA-OEt

Amino Acid Ethyl Ester

AA

Amino Acid

TFA

Trifluoroacetic acid

CHCA

α-cyano-4-hydroxycinnamic acid

DMSO_d6

Deuterated dimethylsulfoxide

TLC

Thin-layer chromatography

DMSO

Dimethylsulfoxide

PBS

Phosphate Buffer Solution

TEA

Triethylamine

Ala

Alanine

Glu

Glutamic Acid

Leu

Leucine

CCDC

Cambridge Crystallographic Data Centre

CSD

Cambridge Structural Database

REU

Rosetta Energy Units

PDB

Protein Data Bank