Cyclic ß2,3-amino acids improve the serum stability of macrocyclic peptide inhibitors targeting the SARS-CoV-2 main protease.

Due to their constrained conformations, cyclic ß2,3-amino acids (cßAA) are key building blocks that can fold peptides into compact and rigid structures, improving peptidase resistance and binding affinity to target proteins, due to their constrained conformations. Although the translation efficiency of cßAAs is generally low, our engineered tRNA, referred to as tRNAPro1E2, enabled efficient incorporation of cßAAs into peptide libraries using the flexible in vitro translation (FIT) system. Here we report on the design and application of a macrocyclic peptide library incorporating 3 kinds of cßAAs: (1R,2S)-2-aminocyclopentane carboxylic acid (ß1), (1S,2S)-2-aminocyclohexane carboxylic acid (ß2), and (1R,2R)-2-aminocyclopentane carboxylic acid. This library was applied to an in vitro selection against the SARS-CoV-2 main protease (Mpro). The resultant peptides, BM3 and BM7, bearing one ß2 and two ß1, exhibited potent inhibitory activities with IC50 values of 40 and 20 nM, respectively. BM3 and BM7 also showed remarkable serum stability with half-lives of 48 and >168 h, respectively. Notably, BM3A and BM7A, wherein the cßAAs were substituted with alanine, lost their inhibitory activities against Mpro and displayed substantially shorter serum half-lives. This observation underscores the significant contribution of cßAA to the activity and stability of peptides. Overall, our results highlight the potential of cßAA in generating potent and highly stable macrocyclic peptides with drug-like properties.

Fine-tuning the tRNA anticodon arm for multiple/consecutive incorporations of ß-amino acids and analogs.

Ribosomal incorporation of ß-amino acids into nascent peptides is much less efficient than that of the canonical α-amino acids. To overcome this, we have engineered a tRNA chimera bearing T-stem of tRNAGlu and D-arm of tRNAPro1, referred to as tRNAPro1E2, which efficiently recruits EF-Tu and EF-P. Using tRNAPro1E2 indeed improved ß-amino acid incorporation. However, multiple/consecutive incorporations of ß-amino acids are still detrimentally poor. Here, we attempted fine-tuning of the anticodon arm of tRNAPro1E2 aiming at further enhancement of ß-amino acid incorporation. By screening various mutations introduced into tRNAPro1E2, C31G39/C28G42 mutation showed an approximately 3-fold enhancement of two consecutive incorporation of ß-homophenylglycine (ßPhg) at CCG codons. The use of this tRNA made it possible for the first time to elongate up to ten consecutive ßPhg's. Since the enhancement effect of anticodon arm mutations differs depending on the codon used for ß-amino acid incorporation, we optimized anticodon arm sequences for five codons (CCG, CAU, CAG, ACU and UGG). Combination of the five optimal tRNAs for these codons made it possible to introduce five different kinds of ß-amino acids and analogs simultaneously into model peptides, including a macrocyclic scaffold. This strategy would enable ribosomal synthesis of libraries of macrocyclic peptides containing multiple ß-amino acids.

Engineering tRNAs for the Ribosomal Translation of Non-proteinogenic Monomers.

Ribosome-dependent protein biosynthesis is an essential cellular process mediated by transfer RNAs (tRNAs). Generally, ribosomally synthesized proteins are limited to the 22 proteinogenic amino acids (pAAs: 20 l-α-amino acids present in the standard genetic code, selenocysteine, and pyrrolysine). However, engineering tRNAs for the ribosomal incorporation of non-proteinogenic monomers (npMs) as building blocks has led to the creation of unique polypeptides with broad applications in cellular biology, material science, spectroscopy, and pharmaceuticals. Ribosomal polymerization of these engineered polypeptides presents a variety of challenges for biochemists, as translation efficiency and fidelity is often insufficient when employing npMs. In this Review, we will focus on the methodologies for engineering tRNAs to overcome these issues and explore recent advances both in vitro and in vivo. These efforts include increasing orthogonality, recruiting essential translation factors, and creation of expanded genetic codes. After our review on the biochemical optimizations of tRNAs, we provide examples of their use in genetic code manipulation, with a focus on the in vitro discovery of bioactive macrocyclic peptides containing npMs. Finally, an analysis of the current state of tRNA engineering is presented, along with existing challenges and future perspectives for the field.

Cheminformatics-Guided Cell-Free Exploration of Peptide Natural Products.

There have been significant advances in the flexibility and power of in vitro cell-free translation systems. The increasing ability to incorporate noncanonical amino acids and complement translation with recombinant enzymes has enabled cell-free production of peptide-based natural products (NPs) and NP-like molecules. We anticipate that many more such compounds and analogs might be accessed in this way. To assess the peptide NP space that is directly accessible to current cell-free technologies, we developed a peptide parsing algorithm that breaks down peptide NPs into building blocks based on ribosomal translation logic. Using the resultant data set, we broadly analyze the biophysical properties of these privileged compounds and perform a retrobiosynthetic analysis to predict which peptide NPs could be directly synthesized in augmented cell-free translation reactions. We then tested these predictions by preparing a library of highly modified peptide NPs. Two macrocyclases, PatG and PCY1, were used to effect the head-to-tail macrocyclization of candidate NPs. This retrobiosynthetic analysis identified a collection of high-priority building blocks that are enriched throughout peptide NPs, yet they had not previously been tested in cell-free translation. To expand the cell-free toolbox into this space, we established, optimized, and characterized the flexizyme-enabled ribosomal incorporation of piperazic acids. Overall, these results demonstrate the feasibility of cell-free translation for peptide NP total synthesis while expanding the limits of the technology. This work provides a novel computational tool for exploration of peptide NP chemical space, that could be expanded in the future to allow design of ribosomal biosynthetic pathways for NPs and NP-like molecules.

De Novo Single-Stranded RNA-Binding Peptides Discovered by Codon-Restricted mRNA Display.

RNA-binding proteins participate in diverse cellular processes, including DNA repair, post-transcriptional modification, and cancer progression through their interactions with RNAs, making them attractive for biotechnological applications. While nature provides an array of naturally occurring RNA-binding proteins, developing de novo RNA-binding peptides remains challenging. In particular, tailoring peptides to target single-stranded RNA with low complexity is difficult due to the inherent structural flexibility of RNA molecules. Here, we developed a codon-restricted mRNA display and identified multiple de novo peptides from a peptide library that bind to poly(C) and poly(A) RNA with KDs ranging from micromolar to submicromolar concentrations. One of the newly identified peptides is capable of binding to the cytosine-rich sequences of the oncogenic Cdk6 3'UTR RNA and MYU lncRNA, with affinity comparable to that of the endogenous binding protein. Hence, we present a novel platform for discovering de novo single-stranded RNA-binding peptides that offer promising avenues for regulating RNA functions.

Macrocyclic Peptides Closed by a Thioether-Bipyridyl Unit That Grants Cell Membrane Permeability.

Membrane permeability is an important factor that determines the virtue of peptides targeting intracellular molecules. By introducing a membrane penetration motif, some peptides exhibit better membrane permeabilities. Previous choices for such motifs have usually been polycationic sequences, but their protease vulnerabilities and modest endosome escapability remain challenging. Here, we report a strategy for macrocyclization of peptides closed by a hydrophobic bipyridyl (BPy) unit, which grants an improvement of their membrane permeability and proteolytic stability compared with the conventional polycationic peptides. We chemically prepared model macrocyclic peptides closed by a thioether-BPy unit and determined their cell membrane permeability, giving 200 nM CP50 (an indicative value of membrane permeability), which is 40-fold better than that of the ordinary thioether macrocycle consisting of the same sequence composition. To discover potent target binders consisting of the BPy unit, we reprogrammed the initiator with chloromethyl-BPy (ClMeBPy) for the peptide library synthesis with a downstream Cys residue(s) and executed RaPID (Random nonstandard Peptide Integrated Discovery) against the bromodomains of BRD4. One of the obtained sequences exhibited a single-digit nanomolar dissociation constant against BRD4 in vitro and showed approximately 2-fold and 10-fold better membrane permeability than positive controls, R9 and Tat peptides, respectively. Moreover, we observed an intracellular activity of the BPy macrocycle tagged with a proteasome target peptide motif (RRRG), resulting in modest but detectable degradation of BRD4. The present demonstration indicates that the combination of the RaPID system with an appropriate hydrophobic unit, such as BPy, would provide a potential approach for devising cell penetrating macrocycles targeting various intracellular proteins.

A comprehensive analysis of translational misdecoding pattern and its implication on genetic code evolution.

The universal genetic code is comprised of 61 sense codons, which are assigned to 20 canonical amino acids. However, the evolutionary basis for the highly conserved mapping between amino acids and their codons remains incompletely understood. A possible selective pressure of evolution would be minimization of deleterious effects caused by misdecoding. Here we comprehensively analyzed the misdecoding pattern of 61 codons against 19 noncognate amino acids where an arbitrary amino acid was omitted, and revealed the following two rules. (i) If the second codon base is U or C, misdecoding is frequently induced by mismatches at the first and/or third base, where any mismatches are widely tolerated; whereas misdecoding with the second-base mismatch is promoted by only U-G or C-A pair formation. (ii) If the second codon base is A or G, misdecoding is promoted by only G-U or U-G pair formation at the first or second position. In addition, evaluation of functional/structural diversities of amino acids revealed that less diverse amino acid sets are assigned at codons that induce more frequent misdecoding, and vice versa, so as to minimize deleterious effects of misdecoding in the modern genetic code.

Translation initiation with exotic amino acids using EF-P-responsive artificial initiator tRNA.

Translation initiation using noncanonical initiator substrates with poor peptidyl donor activities, such as N-acetyl-l-proline (AcPro), induces the N-terminal drop-off-reinitiation event. Thereby, the initiator tRNA drops-off from the ribosome and the translation reinitiates from the second amino acid to yield a truncated peptide lacking the N-terminal initiator substrate. In order to suppress this event for the synthesis of full-length peptides, here we have devised a chimeric initiator tRNA, referred to as tRNAiniP, whose D-arm comprises a recognition motif for EF-P, an elongation factor that accelerates peptide bond formation. We have shown that the use of tRNAiniP and EF-P enhances the incorporation of not only AcPro but also d-amino, ß-amino and γ-amino acids at the N-terminus. By optimizing the translation conditions, e.g. concentrations of translation factors, codon sequence and Shine-Dalgarno sequence, we could achieve complete suppression of the N-terminal drop-off-reinitiation for the exotic amino acids and enhance the expression level of full-length peptide up to 1000-fold compared with the use of the ordinary translation conditions.

In vitro selection of macrocyclic peptide inhibitors containing cyclic γ2,4-amino acids targeting the SARS-CoV-2 main protease.

γ-Amino acids can play important roles in the biological activities of natural products; however, the ribosomal incorporation of γ-amino acids into peptides is challenging. Here we report how a selection campaign employing a non-canonical peptide library containing cyclic γ2,4-amino acids resulted in the discovery of very potent inhibitors of the SARS-CoV-2 main protease (Mpro). Two kinds of cyclic γ2,4-amino acids, cis-3-aminocyclobutane carboxylic acid (γ1) and (1R,3S)-3-aminocyclopentane carboxylic acid (γ2), were ribosomally introduced into a library of thioether-macrocyclic peptides. One resultant potent Mpro inhibitor (half-maximal inhibitory concentration = 50 nM), GM4, comprising 13 residues with γ1 at the fourth position, manifests a 5.2 nM dissociation constant. An Mpro:GM4 complex crystal structure reveals the intact inhibitor spans the substrate binding cleft. The γ1 interacts with the S1' catalytic subsite and contributes to a 12-fold increase in proteolytic stability compared to its alanine-substituted variant. Knowledge of interactions between GM4 and Mpro enabled production of a variant with a 5-fold increase in potency.

Drop-off-reinitiation at the amino termini of nascent peptides and its regulation by IF3, EF-G, and RRF.

In translation initiation in prokaryotes, IF3 recognizes the interaction between the initiator codon of mRNA and the anticodon of fMet-tRNAini and then relocates the fMet-tRNAini to an active position. Here, we have surveyed 328 codon-anticodon combinations for the preference of IF3. At the first and second base of the codon, only Watson-Crick base pairs are tolerated. At the third base, stronger base pairs, for example, Watson-Crick, are more preferred, but other types of base pairs, for example, G/U wobble, are also tolerated; weaker base pairs are excluded by IF3. When the codon-anticodon combinations are unfavorable for IF3 or the concentration of IF3 is too low to recognize any codon-anticodon combinations, IF3 fails to set the P-site fMet-tRNAini at the active position and causes its drop-off from the ribosome. Thereby, translation reinitiation occurs from the second aminoacyl-tRNA at the A site to yield a truncated peptide lacking the amino-terminal fMet. We refer to this event as the amino-terminal drop-off-reinitiation. We also showed that EF-G and RRF are involved in disassembling such an aberrant ribosome complex bearing inactive fMet-tRNAini Thereby EF-G and RRF are able to exclude unfavorable codon-anticodon combinations with weaker base pairs and alleviate the amino-terminal drop-off-reinitiation.

Ribosomal incorporation of negatively charged d-α- and N-methyl-l-α-amino acids enhanced by EF-Sep.

Ribosomal incorporation of d-α-amino acids (dAA) and N-methyl-l-α-amino acids (MeAA) with negatively charged sidechains, such as d-Asp, d-Glu, MeAsp and MeGlu, into nascent peptides is far more inefficient compared to those with neutral or positively charged ones. This is because of low binding affinity of their aminoacyl-transfer RNA (tRNA) to elongation factor-thermo unstable (EF-Tu), a translation factor responsible for accommodation of aminoacyl-tRNA onto ribosome. It is well known that EF-Tu binds to two parts of aminoacyl-tRNA, the amino acid moiety and the T-stem; however, the amino acid binding pocket of EF-Tu bearing Glu and Asp causes electric repulsion against the negatively charged amino acid charged on tRNA. To circumvent this issue, here we adopted two strategies: (i) use of an EF-Tu variant, called EF-Sep, in which the Glu216 and Asp217 residues in EF-Tu are substituted with Asn216 and Gly217, respectively; and (ii) reinforcement of the T-stem affinity using an artificially developed chimeric tRNA, tRNAPro1E2, whose T-stem is derived from Escherichia coli tRNAGlu that has high affinity to EF-Tu. Consequently, we could successfully enhance the incorporation efficiencies of d-Asp, d-Glu, MeAsp and MeGlu and demonstrated for the first time, to our knowledge, ribosomal synthesis of macrocyclic peptides containing multiple d-Asp or MeAsp. This article is part of the theme issue 'Reactivity and mechanism in chemical and synthetic biology'.

In Vitro Selection of Macrocyclic α/ß3-Peptides against Human EGFR.

Here, we report ribosomal construction of thioether-macrocyclic α/ß3-peptide libraries in which ß-homoglycine, ß-homoalanine, ß-homophenylglycine, and ß-homoglutamine are introduced by genetic code reprogramming. The libraries were applied to the RaPID (Random nonstandard Peptides Integrated Discovery) selection against human EGFR to obtain PPI (protein-protein interaction) inhibitors. The resulting peptides contained up to five ß3-amino acid (ß3AA) residues and exhibited outstanding binding affinity, PPI inhibitory activity, and proteolytic stability, which were attributed to the ß3AAs included in the peptides. This showcase work has demonstrated that the use of such ß3AAs enhances the drug-like properties of peptides, providing a unique platform for the discovery of de novo macrocycles against a protein of interest.

In Vitro Genetic Code Reprogramming for the Expansion of Usable Noncanonical Amino Acids.

Genetic code reprogramming has enabled us to ribosomally incorporate various nonproteinogenic amino acids (npAAs) into peptides in vitro. The repertoire of usable npAAs has been expanded to include not only l-α-amino acids with noncanonical sidechains but also those with noncanonical backbones. Despite successful single incorporation of npAAs, multiple and consecutive incorporations often suffer from low efficiency or are even unsuccessful. To overcome this stumbling block, engineering approaches have been used to modify ribosomes, EF-Tu, and tRNAs. Here, we provide an overview of these in vitro methods that are aimed at optimal expansion of the npAA repertoire and their applications for the development of de novo bioactive peptides containing various npAAs.

Drop-off-reinitiation triggered by EF-G-driven mistranslocation and its alleviation by EF-P.

In ribosomal translation, peptidyl transfer occurs between P-site peptidyl-tRNA and A-site aminoacyl-tRNA, followed by translocation of the resulting P-site deacylated-tRNA and A-site peptidyl-tRNA to E and P site, respectively, mediated by EF-G. Here, we report that mistranslocation of P-site peptidyl-tRNA and A-site aminoacyl-tRNA toward E and A site occurs when high concentration of EF-G triggers the migration of two tRNAs prior to completion of peptidyl transfer. Consecutive incorporation of less reactive amino acids, such as Pro and d-Ala, makes peptidyl transfer inefficient and thus induces the mistranslocation event. Consequently, the E-site peptidyl-tRNA drops off from ribosome to give a truncated peptide lacking the C-terminal region. The P-site aminoacyl-tRNA allows for reinitiation of translation upon accommodation of a new aminoacyl-tRNA at A site, leading to synthesis of a truncated peptide lacking the N-terminal region, which we call the 'reinitiated peptide'. We also revealed that such a drop-off-reinitiation event can be alleviated by EF-P that promotes peptidyl transfer of Pro. Moreover, this event takes place both in vitro and in cell, showing that reinitiated peptides during protein synthesis could be accumulated in this pathway in cells.

In Vitro Selection of Foldamer-Like Macrocyclic Peptides Containing 2-Aminobenzoic Acid and 3-Aminothiophene-2-Carboxylic Acid.

Aromatic cyclic ß2,3-amino acids (cßAAs), such as 2-aminobenzoic acid and 3-aminothiophene-2-carboxylic acid, are building blocks that can induce unique folding propensities of peptides. Although their ribosomal elongation had been a formidable task due to the low nucleophilicity of their amino groups, we have recently overcome this issue by means of an engineered tRNAPro1E2 that enhances their incorporation efficiency into nascent peptide chains. Here we report ribosomal synthesis of a random macrocyclic peptide library containing aromatic and aliphatic cßAAs, and its application to de novo discovery of binders against human IFNGR1 and FXIIa as model targets. The potent binding peptides showed not only high inhibitory activity but also high protease resistance in human serum. Moreover, these cßAAs play a critical role in exhibiting their properties, establishing a discovery platform for de novo foldamer-like macrocycles containing such unique building blocks.

In Vitro Selection of Thioether-Closed Macrocyclic Peptide Ligands by Means of the RaPID System.

The Random nonstandard Peptides Integrated Discovery (RaPID) system enables efficient screening of macrocyclic peptides with high affinities against target molecules. Random peptide libraries are prepared by in vitro translation using the Flexible In vitro Translation (FIT) system, which allows for incorporation of diverse nonproteinogenic amino acids into peptides by genetic code reprogramming. By introducing an N-chloroacetyl amino acid at the N-terminus and a Cys at the downstream, macrocyclic peptide libraries can be readily generated via posttranslational thioether formation. Here, we describe how to prepare a thioether-closed macrocyclic peptide library, and its application to the RaPID screening.

Consecutive Ribosomal Incorporation of α-Aminoxy/α-Hydrazino Acids with l/d-Configurations into Nascent Peptide Chains.

α-Aminoxy and α-hydrazino acids are ß-amino acid analogs with ß-carbons replaced by oxygen and nitrogen, respectively. Such heteroatoms dictate the folding of peptides into specific secondary structures called pseudo-γ-turns. Achiral α-aminoxyacetic acid (NOGly) and l-α-hydrazinophenylalanine (l-NNPhe) have been shown to be suitable for single incorporation during ribosomal translation, but whether ribosomes tolerate other types of α-aminoxy/α-hydrazino acids with l/d-configurations is unknown. Moreover, whether multiple or consecutive incorporations are possible remains unclear. We show, for the first time, multiple and consecutive incorporations of α-aminoxy/α-hydrazino acids with l/d-configurations into various model peptides, including macrocyclic peptide scaffolds.

Macrocyclic peptides exhibit antiviral effects against influenza virus HA and prevent pneumonia in animal models.

Most anti-influenza drugs currently used, such as oseltamivir and zanamivir, inhibit the enzymatic activity of neuraminidase. However, neuraminidase inhibitor-resistant viruses have already been identified from various influenza virus isolates. Here, we report the development of a class of macrocyclic peptides that bind the influenza viral envelope protein hemagglutinin, named iHA. Of 28 iHAs examined, iHA-24 and iHA-100 have inhibitory effects on the in vitro replication of a wide range of Group 1 influenza viruses. In particular, iHA-100 bifunctionally inhibits hemagglutinin-mediated adsorption and membrane fusion through binding to the stalk domain of hemagglutinin. Moreover, iHA-100 shows powerful efficacy in inhibiting the growth of highly pathogenic influenza viruses and preventing severe pneumonia at later stages of infection in mouse and non-human primate cynomolgus macaque models. This study shows the potential for developing cyclic peptides that can be produced more efficiently than antibodies and have multiple functions as next-generation, mid-sized biomolecules.

Uniform affinity-tuning of N-methyl-aminoacyl-tRNAs to EF-Tu enhances their multiple incorporation.

In ribosomal translation, the accommodation of aminoacyl-tRNAs into the ribosome is mediated by elongation factor thermo unstable (EF-Tu). The structures of proteinogenic aminoacyl-tRNAs (pAA-tRNAs) are fine-tuned to have uniform binding affinities to EF-Tu in order that all proteinogenic amino acids can be incorporated into the nascent peptide chain with similar efficiencies. Although genetic code reprogramming has enabled the incorporation of non-proteinogenic amino acids (npAAs) into the nascent peptide chain, the incorporation of some npAAs, such as N-methyl-amino acids (MeAAs), is less efficient, especially when MeAAs frequently and/or consecutively appear in a peptide sequence. Such poor incorporation efficiencies can be attributed to inadequate affinities of MeAA-tRNAs to EF-Tu. Taking advantage of flexizymes, here we have experimentally verified that the affinities of MeAA-tRNAs to EF-Tu are indeed weaker than those of pAA-tRNAs. Since the T-stem of tRNA plays a major role in interacting with EF-Tu, we have engineered the T-stem sequence to tune the affinity of MeAA-tRNAs to EF-Tu. The uniform affinity-tuning of the individual pairs has successfully enhanced the incorporation of MeAAs, achieving the incorporation of nine distinct MeAAs into both linear and thioether-macrocyclic peptide scaffolds.

In Vitro Selection of Macrocyclic d/l-Hybrid Peptides against Human EGFR.

d/l-Hybrid peptides are an attractive class of molecular modality because they are able to exhibit high proteolytic stability and unique structural diversity which cannot be accessed by those consisting of only proteinogenic l-amino acids. Despite such an expectation, it has not been possible to devise de novo d/l-hybrid peptides capable of disrupting the function of a protein target(s) due to the lack of an effective method that reliably constructs a highly diverse library and screens active species. Here we report for the first time construction of a library consisting of 1012 members of macrocyclic d/l-hybrid peptides containing five kinds of d-amino acids and performance of the RaPID selection against human EGFR as a showcase to uncover PPI (protein-protein interaction) inhibitors.