{"paper_id":"419c12a3-9a9f-434e-87fa-acb9adfac512","body_text":"In the field of biomimicry,\nscientists find inspiration in nature,\nemulating its native models and solutions to craft molecules and processes\nspecifically tailored to address human challenges. This scientific\npursuit involves building upon foundational knowledge, and strategically\nharnessing inspiration from natural molecular architectures and functional\nmechanisms.  The widespread adoption of\nthe strategy to replicate the structures and functions of peptides\nand proteins is a fundamental pillar in the dynamic landscape of drug\ndesign, discovery, and development. The significance of the field\nis underscored by a robust and continuously expanding body of scientific\nliterature which is reflective of ongoing advancements that shape\nour understanding of these essential biological entities.\nPeptides and proteins are indispensable\ncomponents of cellular\nentities, assuming crucial roles in vital biological processes, providing\nstructural support to cells and tissues, facilitating signal transmission,\nregulating physiological functions, and contributing to the functioning\nof the immune system. The structural complexity of these biomolecules\nspans a spectrum, ranging from small peptides characterized by single\nsecondary structures or random configurations to intricate assemblies\nof helices, sheets, and turns observed in more complex proteins.  This diversity in structural complexity highlights\nthe adaptability and versatility of peptides and proteins as they\nexecute essential functions within the dynamic landscape of cellular\nbiology.\nThe pharmaceutical industry has recognized the therapeutic\npotential\nof peptides in addressing unmet medical needs, positioning them as\na valuable adjunct or even a preferable alternative to small molecules.\nPeptides play a transformative role in modern pharmaceutical research,\nserving as key drivers in the advancement of both biological and chemical\nsciences.  Early 20th-century research\nefforts, which sought to unravel the structures and biological functions\nof peptide hormones like insulin, oxytocin, vasopressin, and gonadotropin-releasing\nhormones ( Table  \n ,\nTable S2, Table S6), have led to numerous innovations in pharmacology,\nchemistry, biology, and technologies fundamental to current drug discovery\nprocesses.  The discovery of insulin in\n1921 marked a transformative milestone. Within a year, it transitioned\nfrom laboratory research to clinical application and subsequently\nemerged as the first commercially available peptide therapy in 1922.\nAnother pivotal moment occurred in 1982 with the introduction of human\ninsulin, exemplified by Eli Lilly and Company’s development\nof Humulin, the inaugural synthetic insulin manufactured through recombinant\nDNA technology. This advancement eventually resulted in discontinuing\nthe original insulin product derived from animals, which had been\nused for six decades ( Figure  \n ).\nInsulin journey: from\nits discovery to the commercialization of\nthe first synthetic human insulin. Created in BioRender.\nSC: subcutaneous, O: orally.\nAlthough peptides have a long history, by the end\nof the 20th century,\nthey had largely been reduced to a niche pharmaceutical category because\nlarge-scale production was prohibitively expensive; thus, only peptide\nhormones effective at low doses were viable on the market. However,\nthe past two decades have seen a significant revival in peptide drug\ndiscovery efforts. Since 2000, numerous noninsulin peptide drugs have\nbeen approved globally, with several achieving substantial market\nsuccess.  Concurrent advancements in recombinant\nbiologics have also led to a renewed interest in peptides, as both\nfields share similar biological properties and scientific developments.\nThese achievements have encouraged pharmaceutical companies to reconsider\nthe potential of peptide drug discovery, resulting in a renewed wave\nof investment in this area.  At present,\naround 140 peptide-based medications are available globally, with\nongoing steady progress in the development of new peptide therapeutics.\n( Figure  \n ). \n , −\nLeft:\nAnnual number of drugs approved by the FDA from 2016 to 2024,\ncategorized as peptides (blue), biologics (orange, including monoclonal\nantibodies, enzymes, and antibody-drug conjugates), and other approved\nentities (green). Right: Percentage of peptides approved relative\nto the total number of approved entities each year. Created in BioRender.\nFDA-approved drugs can be broadly classified into\n2 major groups\nbased on their nature and mode of action: small molecules and biologics.\nSmall-molecule drugs contain up to 100 atoms and are very stable in\nseveral conditions. Biologics are therapeutics that come from living\norganisms and include recombinant proteins, monoclonal antibodies,\ncell therapies, vaccines and genes. Peptide therapeutics occupy a\nunique position in the pharmaceutical landscape, bridging the gap\nbetween small molecules and biologics ( Figure  \n ).\nDrugs are categorized\ninto three main classes based on their size\nand properties: small molecules (100–1000 Da), peptides (1–30\nkDa), and biologics (>30 kDa). Small molecules, such as paracetamol,\nare considered highly druggable, as they can be administered orally\nand exhibit favorable pharmacokinetic profiles. In contrast, peptides\n(in the figure exemplified by insulin having two peptide chains connected\nby disulfide bridges) and biologics, such as antibodies, are less\nstable, less soluble, and more expensive to produce. However, peptides\nand biologics offer exceptional specificity, often demonstrating superior\npharmacological efficacy. Created in BioRender.\nSmall molecules have long dominated the global\ndrug market, benefiting\nfrom advantages such as cost-effectiveness, oral administration, and\nfacile synthesis. Moreover, their inherent ability to penetrate cellular\nmembranes widens the scope of biological targets that can be addressed.\nConversely, peptides face challenges related to proteolytic instability\nand rapid clearance, which impact pharmacokinetic optimization and\nprevent tissue accumulation.  Notably\npeptides offer significant advantages, including higher specificity\nand minimal hepatic metabolism, which is frequently associated with\nsmall-molecule drugs. Additionally, human dosage predictions for peptides\nusing allometric scaling tend to be more straightforward compared\nto small molecules, making dose-ranging studies in clinical trials\neasier to conduct.\nThe future of\nthe peptide field looks promising, as continued scientific\nprogress is set to overcome existing challenges and fully exploit\nthe potent pharmacological potential of peptides in clinical applications\nand beyond.\nPeptidomimetics is a class of compounds designed\nto imitate peptides\nby replicating specific physicochemical properties of certain amino\nacids or isolated secondary structures. Early examples of peptidomimetics\nfocused on mimicking the primary structures of peptide hormones and\nprotease substrates.  In the past decade,\npeptidomimetics have gained recognition as valuable bioactive agents\nand promising drug candidates, especially in the area of targeting\nand modulating protein–protein interactions. This approach\naddresses challenges associated with traditional peptides, including\nlimited enzymatic stability in the gastrointestinal tract and serum,\ninadequate absorption, rapid excretion through hepatic and renal pathways,\ncompromised targeting capabilities arising from the intrinsic rotational\nflexibility of amino acids, and the potential for inducing antigenicity\nand unpredictable immune responses. Key factors like solubility, stability,\nbioavailability, and affinity play critical roles in pharmaceutical\ncompanies’ development of new drugs. Peptidomimetics, by offering\na means to enhance these properties, presents a promising route for\ndeveloping improved medicines. \n ,\nAntibodies have\nproven to be valuable treatments and provided enhanced\nspecificity and significant pharmacokinetic benefits compared to peptides\n( Figure  \n ). However,\nchallenges related to cost, solubility, stability, immunogenicity\nand bioavailability persist when developing these new therapeutics.\nIn contrast, peptides offer advantages in this domain, including ease\nof production, cost-effectiviness, stability and reduced immunogenicity.\nThis review article covers literature\nmainly from 2004 to 2024\nand explores various modifications aimed at synthesizing peptidomimetics.\nA primary focus is placed on enhancing stability, binding affinity,\nand biological activity to develop more efficacious drugs. We systematically\nreview methods for backbone modifications, starting with conservative\napproaches involving minimal alterations at specific positions along\nthe peptide backbone. Various documented techniques for peptide backbone\nmanipulation are explored, each tailored to achieve specific desired\nproperties. The discussion also encompasses the recent advances leveraging\nmodern biotechnological tools to create novel molecular transformations.\nThe final section of the review will pivot toward practical applications\nof these molecules, especially peptides and peptidomimetics, to underscore\nthe successful deployment of these molecules in various areas within\nthe field of medicinal chemistry, medicine and pharmacology. Peptide\napplications in diverse fields, including diabetes, obesity, cancer,\nand infectious disease research, are highlighted. This review evaluates\nboth the successes and limitations within the field to identify areas\nthat require further investigation. Its ultimate goal is to consolidate\nthe current understanding of the discipline and to propose potential\npathways for future progress. Each topic has been investigated to\nthe degree that benefits the broader scientific community. This review\naims to be an accessible review of general interest to the chemistry\ncommunity because it consolidates a vast array of relevant information,\nprovides an in-depth analysis of key concepts, and critically evaluates\nadvancements in the field. By synthesizing and presenting complex\ninformation in an accessible manner, it serves as a valuable reference\nfor researchers, educators, and practitioners, fostering a deeper\nunderstanding and engagement within the chemistry community.\n\nThe optimization of peptide therapeutics necessitates a thorough\nmultiparametric approach that evaluates the effects of each structural\nmodification on key physicochemical properties, including potency,\nselectivity, stability, solubility, pharmacokinetic characteristics,\nand toxicity. Distinct differences exist between the optimization\nprocesses for peptides versus small molecules. Unlike small molecules,\npeptides can maintain high potency throughout the optimization process.\nFurthermore, the polymeric structure of peptides enables precise modifications\nat each residue, often leading to synergistic enhancements in overall\nperformance when multiple local modifications occur.\nNonetheless,\npeptides possess intrinsic limitations concerning\nabsorption, distribution, metabolism, excretion, and toxicity (ADMET)\nprofiles. They typically demonstrate low absorption rates and limited\nplasma distribution and primarily undergo proteolytic metabolism characterized\nby amide bond cleavage in the backbone. Additionally, peptides are\nmainly excreted by kidneys, resulting in comparatively reduced toxicological\nrisks.\nA general outline of the\noptimization process includes ( Figure  \n ): i. Determining the minimum active sequence:\nThe peptide is subjected to iterative truncation of amino acids from\nboth the C- and N-termini to identify the core sequence essential\nfor the desired biological activity. ii. Conducting positional scanning to\nidentify critical residues: Traditionally executed using an  l -alanine scan, positional scanning involves substituting each side\nchain with the smallest alternative while maintaining a similar conformational\nprofile to evaluate its importance for biological activity. Recent\nadvancements in synthetic and purification techniques have enabled\nscanning with a defined set of amino acids that exhibit diverse physical\nproperties. \n iii. Shielding from degradation at the\ntermini: Modifications to the C- and N-termini are performed to inhibit\nthe degradative action of carboxy- and aminopeptidases, respectively.\nCommonly employed analogues include C-terminal primary amides and\nN-terminal acetylation; however, optimization toward unnatural analogues\nmay be necessary if these modifications are not well tolerated. iv. Identifying sites vulnerable\nto proteolysis:\nInitial exploration of the structure–activity relationship\n(SAR), along with pharmacokinetic experiments, stability assays, and\nmetabolite detection, aids in pinpointing proteolytically susceptible\namide bonds within the sequence. \n v. Enhancing proteolytic stability\nthrough\nbackbone modification: Achieving proteolytic stability while retaining\nbiological activity presents a significant challenge in peptide optimization.\nVarious strategies are available for modifying labile amide bonds,\nbut preserving the desired conformation and binding affinity can be\ncomplex. vi. Formulation\ndevelopment: Appropriate\nformulations for the peptide drug are developed to ensure stability,\nease of administration, and patient compliance. This process may involve\nselecting suitable excipients, dosage forms, and delivery routes.\nDetermining the minimum active sequence:\nThe peptide is subjected to iterative truncation of amino acids from\nboth the C- and N-termini to identify the core sequence essential\nfor the desired biological activity.\nConducting positional scanning to\nidentify critical residues: Traditionally executed using an  l -alanine scan, positional scanning involves substituting each side\nchain with the smallest alternative while maintaining a similar conformational\nprofile to evaluate its importance for biological activity. Recent\nadvancements in synthetic and purification techniques have enabled\nscanning with a defined set of amino acids that exhibit diverse physical\nproperties.\nShielding from degradation at the\ntermini: Modifications to the C- and N-termini are performed to inhibit\nthe degradative action of carboxy- and aminopeptidases, respectively.\nCommonly employed analogues include C-terminal primary amides and\nN-terminal acetylation; however, optimization toward unnatural analogues\nmay be necessary if these modifications are not well tolerated.\nIdentifying sites vulnerable\nto proteolysis:\nInitial exploration of the structure–activity relationship\n(SAR), along with pharmacokinetic experiments, stability assays, and\nmetabolite detection, aids in pinpointing proteolytically susceptible\namide bonds within the sequence.\nEnhancing proteolytic stability\nthrough\nbackbone modification: Achieving proteolytic stability while retaining\nbiological activity presents a significant challenge in peptide optimization.\nVarious strategies are available for modifying labile amide bonds,\nbut preserving the desired conformation and binding affinity can be\ncomplex.\nFormulation\ndevelopment: Appropriate\nformulations for the peptide drug are developed to ensure stability,\nease of administration, and patient compliance. This process may involve\nselecting suitable excipients, dosage forms, and delivery routes.\nDevelopment of a peptide drug begins with its discovery and optimization,\nwhich involves a series of chemical modifications to improve its stability,\nactivity, and formulation. The peptide is then evaluated in vitro\nand in animal models to identify its ADMET profile before progressing\nto human studies, which are conducted in four phases known as clinical\ntrials (Phase 0 to Phase 3). To initiate clinical trials, the sponsor\nmust submit an IND (Investigational New Drug) application to the FDA.\nAfter Phases 1, 2, and 3, the sponsor must submit a report to the\nFDA. At the end of Phase 3, the sponsor may submit an NDA (New Drug\nApplication) to request approval for market release, which will be\nevaluated by the FDA. Following commercialization, the drug is monitored\nto assess its therapeutic effects and potential side effects (Phase\n4). While this description focuses on peptides, it is a general workflow\napplicable to the development of any drug. It always includes a research\nand development phase followed by preclinical and clinical studies,\nall of which are evaluated and approved by the FDA. Created in BioRender.\nA mere 20 natural amino acids are crucial in various\nbiological\nfunctions and structural variations. Post-translational modifications\nfurther enhance the intrinsic diversity in peptide and protein structure\nand function. The strategic addition of nonproteinogenic amino acids,\nalong with several synthetic moieties and techniques, allows for a\nsignificant expansion of this diversity.\nThe building blocks employed to modify the peptide backbone\nexhibit\ndiverse structural variations. Some closely resemble native L-α-residues,\nwhile others bear little similarity to conventional protein backbones.\nIn addition to structural diversity, there is considerable variation\nin the number and density of backbone modifications within a given\nchain. These modifications can be localized or can span a significant\nportion of the mimetics, with the option to create mimetics that feature\nan entirely artificial backbone ( Figure  \n ). However, the latter approach is often\nunnecessary and can sometimes be undesirable, as peptides do not require\na complete “transformation” into peptidomimetic forms.  The more common strategy involves localized\nsubstitutions, which typically focus on a single amino acid or a short\nsegment of contiguous backbone or side chains.\nDifferent types of localized\nbackbone modifications include changes\nin stereochemistry (except for glycine that lacks a chiral center),\nsubstitution of backbone atoms (such as in azapeptides, depsipeptides,\nand thiodepsipeptides), shifting of side chains (as in peptoids),\naddition of additional groups (e.g., alkylations), chain elongation\n(e.g., β-amino acids), and both elongation and substitution\n(e.g., oligoureas and peptidosulfonamides). Cyclization represents\nan additional form of backbone modification; however, this topic is\naddressed separately in  Section  \n .\nMoreover, peptide-peptidomimetic hybrids present\na sophisticated\napproach, allowing for the fine-tuning of binding affinity, resistance\nto proteolytic degradation, and clearance rates by balancing peptide\nand peptidomimetic components. Illustrative examples include peptoid-peptide,\npeptidosulfonamide-peptide, and urea peptidomimetic-peptide hybrids. \n −\nBelow, we present the most common strategies for introducing\nlocalized\nmodifications in peptides.\nAlthough\nD-amino acids are infrequently encountered in nature, they assume\nsignificant roles in specific structures and biological activities\n( Figure  \n a). For instance,\nthey are crucial constituents for bacterial cell walls and antibiotics. \n , \n  Simultaneously, D-amino acids are indispensable for the functional\nintegrity of various hormones and neurotransmitters in mammals. \n , \n  A peptide in which the chirality of amino acid residues transitions\nfrom L to D is termed an inverso analogue, representing the perfect\nmirror image of the original sequence. A retro-inverso peptide is\na modified version of a peptide where both the sequence and chirality\nof the original peptide are reversed. The purpose of creating retro-inverso\npeptides is to design more stable peptides that resist degradation\nby proteases.\nMost experimental\nwork involves the generation of peptides in which a single or a few\namino acids are substituted with their corresponding D-version. D-amino\nacid substitutions have been used in several applications, including\ndevelopment of antimicrobial peptides, improvement of the enzymatic\nstability, enhancement of the antiangiogenic activity and anticancer\naction, assistance of the metal-based peptide bond hydrolysis and\ncontrol of the peptide hydrogel degradation in cells. \n −\nInstead of relying on a single or a few point mutations, Schumacher\net al. developed a phage display technique to screen for and identify\nD-peptide ligands that exhibit resistance to proteolytic degradation.  This method involves synthesizing proteins with\nD-amino acids to select peptides from a phage display library that\nexpresses random L-amino acid peptides. Notably, they discovered that\nL-peptides could bind to proteins composed entirely of D-amino acids.\nTheir findings suggested that an all-D-amino acid peptide could bind\neffectively to its natural protein counterpart made up entirely of\nL-amino acids. This implies that inverso analogs of peptides or proteins\nmight replicate the structure and function of the originals, underscoring\ntheir potential as therapeutic molecules. Encouraging outcomes from\nthese trials subsequently prompted the synthesis of more all-D-peptide\nfragments.  Synthetic all-D-peptides have\nfound application as mirror-image molecules in screening libraries\nof nucleic acids or genetically encoded proteins to identify specific\nbinding ligands.  Moreover, D-peptides\nexhibit diverse functions and showcase resistance to proteolytic degradation,\ngarnering considerable attention in the field of drug discovery.\nModifying protein domains, particularly those featuring functionally\nrelevant surface-exposed loops, is also feasible. This is exemplified\nby incorporating a native receptor-binding peptide loop onto a scaffold\nconstructed with bridges of  d -cysteine residues.  Remarkably, this modification does not compromise\ntertiary folding but significantly enhances stability against protease\ndegradation. Etelcalcetide (Parsabiv), a drug composed of a chain\nof seven D-amino acids with a  d -Cys at the N-terminal forming\na disulfide bond with an  l -Cys, was approved for the treatment\nof hyperparathyroidism.\nA study\nconducted by Eberle et al. has demonstrated the potential\nof D-peptides in developing therapies for COVID-19.  Despite the availability of several vaccines for SARS-CoV-2,\nthere remains a critical demand for effective therapeutic options\ndue to the lack of definitive treatments. Rather than targeting the\ninteraction between the spike protein and cellular receptors, this\nstudy aimed to inhibit one of the key proteases involved in viral\nreplication, specifically the 3CL protease, utilizing  d -enantiomeric\npeptide ligands to disrupt its cleavage function. The research showed\nthat a combination of competitive and noncompetitive D-peptides significantly\nenhanced the inhibitory effect on 3CL protease and displayed notable\nresistance to metabolic degradation over an 8-h period.\nSubstituting\nthe α-carbon of an amino acid with nitrogen results in the formation\nof a semicarbazide, leading to the creation of peptides with semicarbazide,\nknown as azapeptides. This modification replaces the rotatable Cα–C­(O)\nbond with a rigid urea Nα–C­(O), significantly altering\nthe chemical and biological properties of the original peptide ( Figure  \n b). Specifically,\nthis substitution eliminates chirality at the α-position, causing\na shift in geometry from tetrahedral to trigonal. Computational and\nstructural analyses reveal that these sequences adopt a β-turn\ngeometry attributed to the planarity of the urea group and lone pair-lone\npair repulsion of the hydrazine.\nAzapeptides exhibit heightened chemical stability and enzymatic resistance\ncompared to amides, making them attractive drug design candidates.  However, Fmoc-protected aza-amino acids, unlike\nnatural amino acids, are either unstable or commercially unavailable.\nConsequently, synthesizing azapeptides necessitates additional steps\nto introduce aza-amino acids into peptide sequences.\nSeveral\nsynthetic pathways are available for incorporating the\naza-amino acid residue. One method leverages hydrazine chemistry and\npeptide coupling, where the side chain of the aza-amino acid is built\non a hydrazine derivative before coupling with a proteogenic amino\nacid ( Scheme  \n a).  Alternatively, this process can be reversed,\nwith the coupling occurring prior to side chain construction on the\naza-residue. While the use of substituted hydrazines in synthesis\ncan be tedious, progress has been made through the “submonomer\napproach”, where diverse side chains can be added to a common\nsemicarbazone intermediate, specifically the semicarbazone-protected\naza-glycine.  The latter method for introducing\nan aza-residue in solid-phase synthesis can be outlined in three steps:\n(a) activation and coupling of the hydrazone, (b) chemoselective deprotonation\nand alkylation of the resulting semicarbazone, and (c) orthogonal\nliberation and aminoacylation of the semicarbazide ( Scheme  \n b). \n , \n  Extending submonomer chemistry beyond N-alkylation, the methodology\nhas been expanded to prepare aza-arylglycines. Various aryl and heteroaryl\niodides, including N-Boc-3-iodoindole and N-trityl-4-iodoimidazole,\nyield aza-indolyl- and aza-imazolylglycine residues, respectively,\nserving as acid-stable mimics of aza-Trp and aza-His. The addition\nof N-Aryl groups onto the semicarbazone is accomplished through Cu­(I)-mediated\nreactions.\na Fmoc-protected hydrazine\nis\nconverted into the corresponding carbazic acid chlorides with phosgene\nat room temperature. The aza-building block is activated to form a\npeptide bond. \n b The hydrazone\n(benzaldehyde hydrazone in the scheme) is activated with  p -nitrophenyl chloroformate to produce carbazate that is coupled to\nthe peptide on a solid support. The product is a semicarbazone. A\nstrong base (potassium  tert -butoxide, KO t Bu) deprotonates the semicarbazone which is alkylated. In this example,\nthe alkylation occurs with benzyl bromide (BnBr). The semicarbazide\nis liberated treating the semicarbazone with hydroxylamine hydrochloride\n(NH 2 OH) in pyridine. Amino acid was activated with diisopropylcarbodiimide\n(DIC) and coupled to the resulting semicarbazide.\nFmoc-protected hydrazine\nis\nconverted into the corresponding carbazic acid chlorides with phosgene\nat room temperature. The aza-building block is activated to form a\npeptide bond.\nThe hydrazone\n(benzaldehyde hydrazone in the scheme) is activated with  p -nitrophenyl chloroformate to produce carbazate that is coupled to\nthe peptide on a solid support. The product is a semicarbazone. A\nstrong base (potassium  tert -butoxide, KO t Bu) deprotonates the semicarbazone which is alkylated. In this example,\nthe alkylation occurs with benzyl bromide (BnBr). The semicarbazide\nis liberated treating the semicarbazone with hydroxylamine hydrochloride\n(NH 2 OH) in pyridine. Amino acid was activated with diisopropylcarbodiimide\n(DIC) and coupled to the resulting semicarbazide.\nSubmonomer strategies have been adapted for other peptidomimetics,\nsuch as peptoids, streamlining the assembly of oligomers. This approach\nproves convenient for constructing libraries of azapeptides on the\nsolid phase, overcoming challenges encountered in synthesis based\non the activation and coupling of N-protected N’-substituted\ncarbazate building blocks, such as oxadiazole formation. Moreover,\nthis approach facilitates the facile addition of diverse functionalizations\nonto aza-residues, providing an avenue to explore various chemical\nreactions, including nucleophilic substitutions, [1,3]-dipolar cycloadditions,\noxidation for pericyclic reactions, and Diels–Alder cycloadditions,\nenabling the incorporation of polar and charged side chains.  While these methods have established a foundation\nfor azapeptides, their applicability for library construction is limited\ndue to the reagents and stringent conditions that may not be compatible\nwith solid-phase peptide synthesis (SPPS).\nJanda proposed leu-enkephalin\nmimetics in the form of pure azapeptides\nor azatides, which involveBoc-protected α-aza-amino acids coupled\nin a linear, stepwise, chain-lengthening fashion.  Despite the existence of synthetic pathways for incorporating\nthe aza-amino acid residue and conducting peptide synthesis, achieving\npure azapeptide synthesis has proven to be a persistent challenge.\nAltiti et al. introduced a new methods based on thiocarbazate building\nblocks as stable precursors for carbonyl-donating reagents and developed\nactivation methods for these thiocarbazates for coupling ( Scheme  \n ). Thiocarbazate\nbuilding blocks are Fmoc-protected aza-amino acids and are easily\nincorporated into both solution-phase and standard SPPS protocols.\nThey group also established protocols for incorporating these activated\naza-amino acids demonostrating the can employ their methodology to\nsystematically modify individual amino acids in peptides via an aza-scan\napproach, generating a small library of aza-amino acid-substituted\npeptide analogues for further biological evaluation.\nAnother strategy to enhance the enzymaric stability of peptides\nis the replacement of a natural α-residue with analogues featuring\na methyl group on the N atom, known as N-Me-α analogues ( Figure  \n d).  N -Methylation of amino acid residues is already present in nature\nand prevalent in nonribosomal peptides.  Notable examples include cyclosporins, miuraenamides, lagunamides,\nand talaropeptides, among others. \n −\nDrawing inspiration from\nthe immunosuppressant cyclosporine with seven  N -methylated\npeptide bonds (Figure S3), selective  N -methylation\nhas been utilized to produce membrane-permeable cyclic peptides overcoming\nsome of the limitations associated with natural peptides by improving\ntheir stability, pharmacokinetic profiles, and overall bioavailability.\nFor many years, multiply  N -methylated peptides\nwere not favored by medicinal chemists due to the challenges associated\nwith their synthesis. The primary obstacle lies in the steric hindrance\nat the  N -methylated site, which complicates amino\nacid coupling. When Wenger achieved the first total synthesis of cyclosporine,\nhe prompted renewed interest in their chemical production.  Wenger carried out the synthesis in solution\nusing Boc chemistry. Fortunately, cyclosporine lacks a diverse array\nof functionalized amino acids, allowing the difficult couplings at\nthe  N -methylated terminus to be accomplished via\nthe formation of a reactive acid chloride.\nMiller and Scanlan\nlater introduced an efficient solid-phase synthesis\napproach, where free amines were activated with an  o -nitrobenzenesulfonyl group, followed by direct methylation using\nmethyl  p -nitrobenzenesulfonate.  Another solid-phase strategy involves the use of preformed  N -methylated building blocks instead of in situ nitrogen\nalkylation.  This method enables fast\nand efficient coupling of  N -methylated amino acids\nthrough a fragmentation approach, using COMU and Oxyma as coupling\nreagents. Additionally,  N -methylated building blocks\nare protected with Alloc rather than Fmoc ( Scheme  \n ). The use of Alloc offers two key advantages:\nfirst, it is less bulky than Fmoc, facilitating coupling; second,\nit can be removed under neutral conditions, minimizing the risk of\ndiketopiperazine (DKP) formation, which can occur when  N -alkyl amino acids are the second residue on a CTC resin. To further\nimprove coupling efficiency, lower-loaded resins are employed. Retratutide\nis a triple glucagon hormone receptor agonist (GLP-1, GIP, and GCGR\nreceptors) and contains an N-Me-Leu residue ( Section  \n ).\na The  N -methyl\namino acid comes with Alloc to protect the amino functionality instead\nof Fmoc. The coupling occurs like a common SPPS with COMU and oxyma\nas activating reagents. Alloc is removed with tetrakis­(triphenylphosphine)­palladium(0),\nPd­(PPh 3 ) 4  in the scheme, and phenylsilane (PhSiH 3 ). After Alloc removal, the amino group is free and the following\namino acid can couple as normally.\nThe  N -methyl\namino acid comes with Alloc to protect the amino functionality instead\nof Fmoc. The coupling occurs like a common SPPS with COMU and oxyma\nas activating reagents. Alloc is removed with tetrakis­(triphenylphosphine)­palladium(0),\nPd­(PPh 3 ) 4  in the scheme, and phenylsilane (PhSiH 3 ). After Alloc removal, the amino group is free and the following\namino acid can couple as normally.\nAdvancements\nin genetic engineering have enabled the incorporation\nof  N -methyl amino acids into peptides or proteins\nby expanding the genetic code. This involves reassigning specific\ncodons to encode  N -methyl amino acids, allowing for\ntheir incorporation by engineering tRNA (tRNA) during ribosomal protein\nsynthesis. Through directed evolution and rational design, orthogonal\ntRNA–synthetase pairs have been developed to specifically recognize  N -methyl amino acids and incorporate them into growing polypeptides.\nThrough genetic engineering, the genetic code can be significantly\nexpanded, allowing the incorporation of many non-natural amino acids\nbeyond just  N -methylated residues. In this section\nof the review, we have primarily focused on organic synthesis, while\na separate paragraph is dedicated to biotechnological methods.\nHowever, replacing the amide hydrogen with a methyl group disrupts\nboth intramolecular and intermolecular hydrogen bonding, which may\nsffect the stabilization of bioactive conformations and recognition\nby receptors.  in addition, the presence\nof the alkyl group on the nitrogen lowers the energy barrier to switch\nbetween  cis  and  trans  configuration\nstabilizing the  cis  configuration.\nModifications are not limited to N-alkylation; inspired\nby nonribosomal\npeptide natural products, researchers are also developing peptides\nfeaturing N-amino (hydrazide) and  N -hydroxy (hydroxamate)\ngroups, characterized by NH 2  or OH substituents on the\nbackbone amide. These modifications have found applications in the\ndevelopment of optimally constrained folds and modulators of protein–protein\ninteractions. Readers interested in a more detailed discussion are\nreferred to the review by Angera et al.\nAn alternative strategy involves attempts to rigidify amino\nacids\nby constraining the φ and ψ angles. For instance, proline\ninherently exhibits rigidity as its φ angle is constrained within\na five-membered ring formed by the Cα-N bond. Over the years,\nvarious sophisticated approaches have been described to achieve such\nangle constraints, with well-known methods including the preparation\nof Freidinger lactam, spirolactams, and α,α-dialkylated\namino acids, particularly α-methyl derivatives, which have been\nextensively studied. \n −\nα,α-Dimethyl amino acids ( Figure  \n d), such as 2-aminoisobutyric acid or Aib,\nhave been explored in various applications. For instance, they imparts\nincreased stabilization, favoring the formation of a 3 10  helix, which is more compact than the typical α-helix, featuring\nthree amino acids per turn instead of the conventional 3.6 residues\nper turn.  Natural peptides containing\ndialkylated amino acids, exemplified by alamethicin, a membrane-channel-forming\npeptide with several Aib residues, further underscore the practicality\nof these modifications.\nIn other examples, leveraging the replacement\nof one or two α-residues\nwith rigidified Cα-methyl analogs has been instrumental in controlling\ndynamics, particularly in the study of intrinsically disordered sequences\nsuch as the activation domain from the p160 transcriptional coactivator\nfor thyroid hormone and retinoid receptors.  Similarly to N-Me-α analogs, α,α-dialkylated amino\nacids obstruct protease binding, making peptides more stable and better\nsuited for use as drugs. Although α,α-dialkylated amino\nacids are now readily available commercially, they may present synthetic\nchallenges due to increased steric hindrance requiring optimization\nof peptide synthesis protocols.  By employing\nDIC, Oxyma, and microwave-assisted synthesis, it has been possible\nto synthesize a sequence of 17 consecutive Aib residues, leading to\nthe first total synthesis of cephibol D, an antifungal peptide.  Both semaglutide and tirzepatide contain Aib\nresidues in their structures to improve their stability against proteases\n( Section  \n ).\nPeptoids,\na class of compounds pioneered by Zuckermann and colleagues, represent\nstructural isomers of natural peptides wherein the side chains are\nshifted from the α-carbon atom to the amide nitrogen atom ( Figure  \n c).  Notably, the side chains in peptoids, except for proline,\nare attached to the nitrogen, rendering peptoid monomers achiral.\nCompared to their peptide counterparts, peptoids exhibit significantly\nlower susceptibility to proteolytic degradation. Given this advantageous\nfeature, it is not surprising that potential applications of peptoids\nhave been extensively reviewed. \n −\nWhile Zuckermann is credited with the formal\ndiscovery and development of peptoids, Bartlett and his colleagues\nhad earlier explored the concept of N-substituted glycine derivatives\nas peptide mimics. They utilized N-(1-phenylethyl)-Gly and N-(methylimidazole)-Gly\nas monomers, mimicking phenylalanine and histidine, to construct a\ncombinatorial array of N-substituted glycine oligomers.  Zuckermann later formalized the concept of peptoids\nand developed a solid-phase synthesis strategy to rapidly generate\npeptoid libraries for drug discovery.\nEfforts to synthesize peptoid oligomers using the established\nSPPS\nmethod, with a preprepared set of Fmoc-protected monomers, presented\nchallenges. This was due to the hindered nature of the secondary amine\nat the growing N-terminus in peptoid chains which leads to slower\ncoupling reactions compared to the primary amines commonly found in\nSPPS. To address this limitation, Zuckermann developed a more efficient\nsynthetic method for obtaining peptoids on solid phase, known as the\n“submonomer method”.  This\ntechnique involves alternating acylation with bromoacetic acid and\nN,N-diisopropyl carbodiimide (DIC), along with nucleophilic displacement\nreactions of the bromide using primary amines ( Scheme  \n ).\na The synthesis on\nsolid phase\nproceeds with iteration of acylation and nucleophilic substitution\nand produces polypeptoids.\nThe synthesis on\nsolid phase\nproceeds with iteration of acylation and nucleophilic substitution\nand produces polypeptoids.\nPeptoids hold significant\nrelevance in the field of antimicrobial\nresistance. Extensive efforts have been directed toward developing\npeptoids that can mimic antimicrobial peptides (AMPs) with the aim\nof improving the poor pharmacokinetic profiles of the latter. The\ncreation of peptide-peptoid hybrids has shown promising results in\ncombating antibiotic-resistant bacterial pathogens, including  Staphylococcus pneumonia, Pseudomonas aeruginosa ,  Escherichia coli ,  Staphylococcus aureus , and  Bacillus subtilis . \n , \n  In addition, peptoids can be used to create advanced materials exploting\ntheir ability to fold and self-assemble and build large combinatorial\nlibraries to identify protein ligands. \n ,\nThe creation of\nnovel peptidic oligomers, distinguished by a wide array of constitutional\nand configurational isomers, is accomplished by introducing additional\natoms between the carboxyl and amino groups of amino acids. Seebach\nand Gellman were at the forefront of synthesizing extended peptides\nmainly derived from β ( Figure  \n e) or γ-amino acids, demonstrating their ability\nto adopt secondary structures such as helices, sheets, and turns,\nthereby exhibiting “protein-like” behavior and opening\nthe research area of “foldamers”. \n − \n \n \n \n  These sequences not only showed resilience to proteolytic\nenzymes but also displayed enhanced pharmacokinetic properties. This\nhas proven effective when applied strategically within α-helices,\nas demonstrated by the periodic substitution of α-amino acid\nresidues with corresponding β-amino acid residues in the parathyroid\nhormone inverse agonist, PTH(7–34).  This modification resulted in the analogue peptide α/β-PTH(7–34),\nwhich preserves the antagonist and inverse agonist activities of the\noriginal α-peptide while exhibiting increased stability against\naggressive proteolytic enzymes. These outcomes highlight the potential\nof PTH-derived peptides with backbone modifications as valuable tools\nfor examining the mechanisms of PTH metabolism and offer new prospects\nfor developing therapeutics aimed at conditions driven by abnormal\nligand-dependent or ligand-independent activity of PTHR1.\nThe\nadditional carbon–carbon bond in β-amino acids increases\nthe flexibility of the peptide bond, which can be a disadvantage when\ndesigning peptide drugs that need to bind to a specific protein site.\nHowever, when the β-carbon (the carbon adjacent to the nitrogen)\nin backbone-extended amino acids is replaced with oxygen, a more rigid\nconformation is observed ( Figure  \n a). This rigidity is due to the lone-pair electron\nrepulsion in the N–O bond. Due to this stability, α-aminoxy\nacids hold potential in peptide drug design. The lone-pair repulsion\nbetween the heteroatoms in α-amino acids leads to the formation\nof a stable eight-membered-ring hydrogen bond between the amino acid\nand adjacent residues, known as the N–O turn ( Figure  \n b).\na)\nStructures of the different amino acids with extended backbone.\nb) the N–O turn formed with aminoxy acids and c) the N–N\nturn formed instead in the presence of hydrazine acids.\nYang et al. demonstrated that oligomers of α-aminoxy\nacids\ncan form a highly stable 8-helix structure, facilitated by the N–O\nturn.  This helical stability can be utilized\nto develop cell-penetrating peptides. Specifically, a hybrid peptide\ncomposed of D-α-aminoxy acids and L-α-amino acids has\nbeen shown to cross cell membranes through direct translocation.  In contrast, β-aminoxy acids, possessing\nan additional carbon atom compared to α-aminoxy acids, exhibit\nmore flexible structures due to the diversity of backbone extensions\nand substitution possibilities. \n ,\nThe replacement\nof the β-carbon nitrogen introduced another\ncategory of peptidomimetics known as hydrazino acids ( Figure  \n a). In peptides containing\nhydrazino acids, the repulsion of lone electron pairs imparts rigidity\nand promotes an intramolecular hydrogen bonding pattern that facilitates\nunique turns. Specifically, in aza-β3-amino acids (hydrazino\nacids with an alkyl substituent on the extra nitrogen atom), a bifurcated\nintramolecular hydrogen bond forms between the carbonyl acceptor (CO i ) and the nitrogen donor (NH i+2 ), creating an eight-membered\nring. The hydrogen bonding interaction is further stabilized by the\nlone pair participation of neighboring nitrogen atoms (N i+1 ).  This structure is referred to as\nthe hydrazino turn or N–N turn ( Figure  \n c). Due to the rapid pyramidal inversion\nof the nitrogen, N–N turns are less rigid than N–O turns.\nHowever, when the aza-β3-amino acid is part of a small ring,\nits configuration and chirality are preserved.\nAza-β3-amino acids can be synthesized from\nN α -substituted-N β -protected hydrazine\nand esters of\nbromoacetate ( Scheme  \n ). However, this reaction typically yields a low output (36–50%).\nAn alternative approach involves the reductive amination of glyoxylic\nacid with N α -substituted-N β -protected\nhydrazine to obtain the desired amino acid.\nHydrazino-based peptidomimetics have shown promising\nbiological\nactivities, such as acting as protease inhibitors and antimicrobial\nmolecules. \n ,\nSuga and colleagues developed\na biotechnological method to synthesize\na peptide library on ribosomes that includes both α-aminoxy\nand α-hydrazino acids. Since β-amino acids are much less\neffective substrates for ribosomal peptide synthesis compared to α-amino\nacids, consecutive elongation is particularly challenging. Their work\nsuccessfully demonstrated the incorporation of α-aminoxyacetic\nacid and L-α-hydrazinophenylalanine during ribosomal translation\nusing the tRNAPro1E2/EF-P system.\nUrea-based peptidomimetics, also known as oligoureas, represent\na class of mimetics wherein a nitrogen moiety replaces the α-carbon\nof γ-amino acid residues ( Figure  \n f). Oligomeric structures composed of repeating urea\nlinkages are named N,N′-linked oligoureas. Due to the presence\nof two NH groups per urea unit, these oligomers form a stronger yet\ntunable hydrogen bonding network in diverse compound classes, including\nbiologically active and self-assembling molecules.\nWithin this\ncategory, aliphatic N,N′-linked oligoureas fall under the foldamer\nfamily. These oligoureas show a strong tendency to form stable helical\nstructures in aqueous environments, establishing a predictable relationship\nbetween the primary sequence and the specific arrangement of side\nchains along the helix. The canonical oligourea helix, which features\n2.5 residues per turn, presents a side chain configuration that, when\nviewed from above, resembles a five-pointed star spanning two turns.\nGuichard and co-workers have shown the ability of oligoureas to substitute\nthe α-helix in a zinc finger domain.  In fact, these mimetics adopt a native-like conformation featuring\na metal-binding site, allowing them to interact with double-stranded\nDNA. The interaction is primarily facilitated by contacts with the\nsubstituted α-helix in the original protein, underscoring the\neffective structural mimicry of this protein segment.\nStrategies\nfor synthesizing urea peptidomimetics have been developed\nusing solid-phase methods, incorporating both Boc and Fmoc protection\nstrategies. \n ,\nPeptidosulfonamides are\nanother class of peptidomimetics where\na sulfonamide replaces the carbonyl amide. However, the S–N\nbond in sulfonamides is not as strong as the amide bond. In fact,\nit can be unstable and undergo hydrolysis under acidic or basic conditions.\nThis instability occurs because sulfonamides lack the resonance stabilization\nfound in peptide bonds. Introducing an additional −CH2–\ngroup results in the preparation of aminoethanesulfonic acid building\nblocks, which help to achieve stable derivatives that are resistant\nto fragmentation ( Figure  \n f).\nLarge-scale synthesis\nof β-substituted aminoethanesulfonic\nacid building blocks is feasible, and these are utilized in the assembly\nof β-peptidosulfonamides. \n , \n  As the β-aminoethane\nsulfonamide residues act as potent helix or β-strand disruptors,\nthese oligomeric peptidomimetics exhibit relatively high flexibility\nand do not adopt well-defined structures.\nAnother approach involves substituting particular\nchemical moieties with isosteresentities that possess similar\nelectronic distributions and physical properties. This local modification\nprimarily focuses on single amino acids and includes replacements\nof backbone, side chain, and dipeptide isosteres. In this section,\nwe focus on the peptide bond isosteres.\nVarious peptide bond\nisosteres have been reported, providing numerous options for enhancing\nproteolytic stability and biological activity.  A noteworthy subset involves replacing the amino functionality\nwith an isosteric atom, such as oxygen (resulting in depsipeptides)\nor sulfur (yielding thiodepsipeptides) ( Figure  \n g). These modifications significantly influence\nthe secondary structure and folding properties of peptides by altering\nhydrogen-bonding patterns.\nDepsipeptides,\npresent in nature and isolated from various microorganisms\nlike bacteria and fungi, manifest diverse antimicrobial activities\nwith a broad spectrum of action. Notable examples include valinomycin,\nwhich functions as a potassium-selective pore, and nonactin, which\nselectively acts as a pore for ammonium. \n ,\nSince the initial identification of natural depsipeptides,\nnumerous\nmethodologies for synthesizing their synthetic counterparts have been\ndocumented.  Ester bonds can be formed\nnot only on the backbone but also on the side chains, utilizing the\nOH side groups of serines and threonines. If the ester bond is formed\non the backbone, the OH group should be added first. Common methods\nto achieve this involve activating the carboxylic acid group and then\nreacting it with α-hydroxy acids.  This esterification process can be carried out using various coupling\nmethods, such as DIC/DMAP, PyBroP/DIEA, and  N -Hydroxysuccinimide. \n − \n \n \n  Additionally, Mitsunobu esterification offers an alternative approach\nby activating the alcohol rather than the carboxylic acid.\nDepsipeptides are highly effective therapeutics,\nparticularly against\ninfections. Beyond their use in therapeutics, depsipeptides serve\nas excellent peptidomimetics with a range of applications. For example,\ndepsipeptides made with Ser or Thr linked to the peptide backbone\nvia an ester bond can function as peptide switches. These peptide\nswitches have been utilized to functionalize alginate hydrogels, where\nthey rearrange upon enzymatic cleavage to expose the YIGSR sequence,\nwhich binds to integrins on the cell membrane.\nAdditionally, depsipeptides are valuable in the synthesis\nof “difficult\npeptides” due to their ability to disrupt the continuity of\nhydrogen bonds in the peptide backbone, preventing aggregation during\npeptide synthesis. In this strategy, the amino acid following Ser\nor Thr is not attached to their N-terminus but to their side chain,\nwhere the OH group can be selectively removed during SPPS ( Scheme  \n ). The coupling performed\nwith common activating reagents results in the formation of an ester\nbond. The following amino acid is then coupled to the amino function\nof Ser or Thr after Fmoc deprotection. Once the peptide synthesis\nis complete and the peptide is cleaved, mild basic aqueous conditions\npromote the O→N shift, resulting in a classic peptide bond\nand the release of the free Ser or Thr.\nThiodepsipeptides are naturally occurring compounds\nformed through\na thioesterification process, in which a cysteine thiol group reacts\nwith the carboxylic group of amino acids or hydroxy acids. An example\nis the macrocyclic thiodepsipeptide thiocoraline, a potent antitumor\nagent isolated from  Micromonospora sp . and  Verrucosispora sp .\nThiodepsipeptides\ncan be synthesized in a similar manner, but instead\nof using Ser or Thr, Cys is incorporated. At mild basic pH, the S→N\nrearrangement is also observed. Although thiodepsipeptides are less\nstable than depsipeptides, making them less suitable for drug development,\nthey are highly valuable as intermediates in chemical synthesis.\nWhile some therapeutic proteins, such as\nantibodies, inherently possess extended half-lives, many endogenous\nmolecules, including peptide hormones, are susceptible to enzymatic\ndegradation, renal clearance, and rapid receptor-mediated elimination,\nresulting in a short plasma half-life (hereafter referred to as ‘half-life’).\nConsequently, considerable research efforts have been focused on developing\ndiverse strategies and technologies aimed at prolonging the half-lives\nof peptides.\nPolyethylene glycol\n(PEG) conjugation, commonly referred to as PEGylation, increases the\nhydrodynamic volume of the conjugate, which helps to prevent renal\nclearance and enhances the pharmacokinetic profiles of biopharmaceuticals.  Glycoengineering, or the conjugation of biopharmaceuticals\nto complex carbohydrates, has also garnered significant interest over\nthe years.  Both strategies often result\nin heterogeneous products due to the polydisperse nature of PEG and\ncarbohydrate polymers. \n ,\nSimilarly, prolongation\nof the half-life can be achieved by fusing\nor conjugating a peptide or protein with hydrophilic peptides, such\nas XTEN, unstructured biodegradable peptides, or sequences rich in\nPro, Ala, and Ser amino acids (PAS tail). \n , \n  XTEN, genetically fused to the biopharmaceuticals, 864-amino acid\npeptide, is rich in Ala, Gly, Glu, Pro, Ser, and Thr residues. It\nis highly soluble, lacks a defined secondary structure, and has a\nlow tendency to aggregate. Additionally, shorter XTEN variants have\nbeen studied for various applications. \n , \n  The PAS tail\nis also highly soluble in physiological solutions and adopts a random\ncoil conformation, demonstrating excellent stability in plasma.  In contrast to PEGylation, XTENylation and\nPASylation produce a more homogeneous product.\nThe versatility\nof peptide conjugation extends across various fields,\nincluding biomedical research, drug development, diagnostics, and\ntherapeutics. In drug discovery, conjugation can involve a pharmacologically\nactive peptide combined with another active molecule to modify the\npharmacokinetic properties, or a peptide may be utilized as a targeting\nor transmembrane delivery vehicle. The direct conjugation of compatible\nactive agents presents significant advantages in clinical development,\npositioning peptides as promising candidates for this purpose. Peptides\ncan be conjugated to a variety of molecules, including ( Figure  \n ): Small molecules and imaging agents: Peptides can be\ntethered to drugs, imaging agents, fluorophores, radiotracers, contrast\nagents, toxins, or chelating agents, serving purposes in therapeutics,\ndiagnostics, and research. \n − \n \n \n Antibodies and antibody fragments: These facilitate\ntargeted drug delivery, imaging, or immunotherapy by directing them\nto specific cells expressing the corresponding antigen. \n , \n \n Lipids: They enhance cellular uptake,\nmembrane insertion\nor stability, and are applicable for drug delivery, cell-penetrating\npeptides (CPPs), or membrane-targeting peptides. \n − \n \n \n Nucleic acids: Efficacy for gene regulation\nor therapy. \n Nanoparticles: They improve targeting, cellular uptake,\nor controlled release properties for drug delivery, imaging, or diagnostics.\nSmall molecules and imaging agents: Peptides can be\ntethered to drugs, imaging agents, fluorophores, radiotracers, contrast\nagents, toxins, or chelating agents, serving purposes in therapeutics,\ndiagnostics, and research. \n −\nAntibodies and antibody fragments: These facilitate\ntargeted drug delivery, imaging, or immunotherapy by directing them\nto specific cells expressing the corresponding antigen. \n ,\nLipids: They enhance cellular uptake,\nmembrane insertion\nor stability, and are applicable for drug delivery, cell-penetrating\npeptides (CPPs), or membrane-targeting peptides. \n −\nNucleic acids: Efficacy for gene regulation\nor therapy.\nNanoparticles: They improve targeting, cellular uptake,\nor controlled release properties for drug delivery, imaging, or diagnostics.\nA peptide drug can be conjugated to various molecules to enhance\nproperties such as enzymatic stability, plasma half-life, and target\nspecificity. Common conjugates utilized in the pharmaceutical field\ninclude biodegradable entities such as lipids, proteins, glycans,\nantibodies, and nucleotides, as well as inorganic compounds like metal\nnanoparticles and chelated metals. Created in BioRender.\nNoncovalent binding with albumin has been effectively\nemployed\nto extend the half-life of peptides by attaching various ligands,\nsuch as fatty acids or antibody domains targeting albumin, known as\nAlbudAb, to the peptides or miniproteins.  Because albumin is abundantly present and possesses a natural ability\nto carry fatty acids as reversible ligands, harnessing this characteristic\nhighlights the potential of fatty acid derivatization and subsequent\nbinding to albumin in prolonging the action profile of peptide drugs.\nAs a result of these advancements, once-daily and, more recently,\nonce-weekly formulations of several peptide drugs currently on the\nmarket or in development have been achieved (see antidiabetic peptides\nin  Section  \n ).\nThese formulations have largely addressed the therapeutic challenges\nassociated with short half-lives of peptide drugs.  Once-weekly therapies offer significant advantages, including\nenhanced convenience, improved treatment adherence, and better health-related\nquality of life. They also contribute to a reduced sense of burden\nassociated with managing chronic conditions.\nThe process of\nfatty acid derivatization, commonly referred to\nas “lipidation”, has undergone meticulous optimization\nto tailor peptide-based or protein-based therapeutics with precise\nmodifications.  It has been theorized\nthat subcutaneously administered peptide drugs, upon undergoing fatty\nacid derivatization, may exhibit prolonged retention at the injection\nsite compared to their nonderivatized counterparts. The retention\nmechanism in this context likely involves interactions between the\nfatty acid side chain and albumin located at the injection site. Consequently,\nit is expected that the rate of diffusion within the tissue postinjection,\nas well as the transit across the capillary wall, will be reduced\ndue to the increased molecular size of the albumin-peptide complex.\nFurthermore, it has been postulated that\nfatty acid derivatization\nmay enhance the self-association of peptide drugs by promoting hydrophobic\ninteractions among peptide monomers.  This increased self-association would result in diminished absorption\nrates from the subcutaneous tissue, as the larger aggregates would\nhave a greater molecular size than the monomers, thus leading to slower\ndiffusion through the tissue and across the capillary wall. In circulation,\nthe larger size of the complexes could protect the bound peptide from\nrenal clearance and reduce the rate of distribution to extravascular\ncompartments.\nChemically, the fatty acid derivatization of target\nmolecules can\nbe accomplished through several methods, including the direct coupling\nof fatty acids to the peptide backbone or via a linker and/or spacer.\nThe linker, which connects the fatty acid to the spacer or the peptide\nbackbone, plays a pivotal role in modulating binding affinity for\nthe target receptor.  The presence of\na spacer can influence receptor binding; longer spacers may mitigate\nthe negative effects on receptor interactions, while shorter spacers\nor the absence of a spacer can provide protection for the peptide\nagainst degradation, thereby contributing to longer half-lives. \n ,\nFatty acids, particularly those characterized by long alkyl\nchains\nwith a single carboxylate group at the distal end (monoacids), have\nbeen well-documented for their strong affinity for binding to albumin,\nwith binding strength correlating positively with the length of the\nalkyl chain.  Initially, fatty monoacids\nwere used to mimic the transport of endogenous fatty acids, followed\nby the introduction of fatty diacids, which provide enhanced affinity\nfor albumin and subsequently longer half-lives due to the presence\nof an additional carboxylic group at the end of the alkyl chain. The\naffinity for albumin is positively correlated with the length of the\nfatty monoacid or diacid. Among the fatty acids tested, 1,18-octadecanedioic\nacid (C 18  diacid) and 1,20-eicosanedioic acid (C 20  diacid) exhibited the highest binding affinities for albumin. \n , \n  The increased hydrophobicity of fatty monoacids compared to fatty\ndiacids affects the solubility, receptor pharmacology, and biophysical\nproperties of the derivatized molecule. Moreover, the derivatization\nof peptides with fatty monoacids enhances their association with cell\nmembranes, promoting internalizationan attribute that has\nbeen recognized for decades.  In contrast,\nfatty diacids possess an additional carboxylic group that enhances\ntheir solubility, resulting in fatty diacid-derivatized peptides being\nless likely to associate with cell membranes and undergo internalization.\nConsequently, incorporating fatty diacids generally aids in maintaining  in vivo  efficacy, as the target peptide or protein is less\nsusceptible to loss due to hydrophobic interactions with cellular\nsurfaces.\nAnother of the most effective and widely utilized\nstrategies for\nextending the half-life of therapeutic proteins or peptides involves\nfusing or conjugating them with the Fc domain of immunoglobulin G\n(IgG) or with albumin. This approach increases the oral administration\nand molecular size of the therapeutic agent, resulting in reduced\nrenal clearance and enhanced half-life due to cellular recycling mediated\nby the neonatal Fc receptor (FcRn).  At\nphysiological pH, FcRn exhibits a low binding affinity for albumin\nand IgG at the cell surface. However, upon internalization of the\ncomplex, the binding affinity of FcRn for both proteins increases\nwithin the acidified environment of endosomes, thereby protecting\nthem from lysosomal degradation. As a result, albumin and IgG are\nrecycled and released from FcRn at the cell surface, thereby extending\ntheir circulation time in the bloodstream.  Moreover, engineering modifications in both the Fc domain and albumin\nto enhance their binding affinity to FcRn at a pH of 6 provide opportunities\nto further extend their half-lives beyond those of native Fc and albumin. \n , \n  These modifications can optimize the therapeutic efficacy of proteins\nand peptides by promoting sustained circulation and improved pharmacokinetic\nprofiles  in vivo .\nNumerous\ncoupling reactions exist to facilitate the connection of these molecules,\nwith several examples provided below ( Scheme  \n ).\na CuAAC (copper-catalyzed\nazide–alkyne\ncycloaddition) and RuAAC (ruthenium-catalyzed azide–alkyne\ncycloaddition) are metal-catalyzed reactions that yield two constitutional\nisomers. In contrast, SPAAC (strain-promoted azide–alkyne cycloaddition)\ninvolves the reaction of an azide group with a strained alkyne without\nthe need for metal catalysis, utilizing the ring strain of cyclooctyne\nas the driving force. Unlike CuAAC, RuAAC, and SPAAC, which involve\na triple bond, thiol-based reactions instead involve the addition\nof a sulfur atom to a double bond.\nCuAAC (copper-catalyzed\nazide–alkyne\ncycloaddition) and RuAAC (ruthenium-catalyzed azide–alkyne\ncycloaddition) are metal-catalyzed reactions that yield two constitutional\nisomers. In contrast, SPAAC (strain-promoted azide–alkyne cycloaddition)\ninvolves the reaction of an azide group with a strained alkyne without\nthe need for metal catalysis, utilizing the ring strain of cyclooctyne\nas the driving force. Unlike CuAAC, RuAAC, and SPAAC, which involve\na triple bond, thiol-based reactions instead involve the addition\nof a sulfur atom to a double bond.\nThe copper-catalyzed\nazide–alkyne cycloaddition (CuAAC)\nis a prominent method due to its high efficiency and selectivity.\nThis reaction involves the incorporation of azide and alkyne functional\ngroups into the peptide and its counterpart, enabling the conjugation\nof the respective molecules to form a triazole linkage. First reported\nin 2002 by K. Barry Sharpless, Valery Fokin and Morten Meldal, this\ncopper­(I)-catalyzed cycloaddition connects azides and terminal alkynes\nto produce 1,4-regioisomers of 1,2,3-triazoles as the sole products. \n , \n  The advancement of using a copper catalyst in aqueous environments\nimproved upon the initial methodology introduced by Rolf Huisgen in\nthe 1970s, which required elevated temperatures.  While commercial sources of copper­(I), such as cuprous\nbromide or iodide, can be utilized, the reaction is notably more effective\nwhen conducted with a combination of copper­(II) and a reducing agent\n(e.g., sodium ascorbate) to generate Cu­(I) in situ. Given the instability\nof Cu­(I) in aqueous solvents, employing stabilizing ligands, such\nas tris­(benzyltriazolylmethyl)­amine (TBTA), enhances the reaction\nyield.  The CuAAC reaction can be performed\nin a range of solvents, including mixtures of water and miscible organic\nsolvents like alcohols, DMSO, DMF, and THF, while acetonitrile is\ntypically avoided due to its strong coordinating ability toward Cu­(I).\nAdditionally, the starting reagents can often be only partially soluble\nfor the reaction to proceed successfully, and in many instances, the\nproduct can be isolated simply through filtration, eliminating the\nneed for extensive purification steps.\nIn contrast, the ruthenium-catalyzed 1,3-dipolar azide–alkyne\ncycloaddition (RuAAC) accommodates both terminal and internal alkynes,\nresulting in the formation of 1,5-disubstituted and 1,4,5-trisubstituted-1,2,3-triazoles.\nUnlike CuAAC, which is limited to terminal alkynes, RuAAC expands\nthe scope by allowing both terminal and internal alkynes to participate\nin the reaction.  Another notable development\nis the discovery of a broad-spectrum silver­(I)-catalyzed azide–alkyne\ncycloaddition reaction (Ag-AAC), which yields 1,4-triazoles. The mechanistic\ndetails of AgAAC closely resemble those of the copper­(I)-catalyzed\nprocess. It is important to note that silver­(I) salts alone are insufficient\nto facilitate cycloaddition; however, the presence of ligated Ag­(I)\nsources significantly enhances the effectiveness of the AgAAC reaction.\nBioorthogonal chemistry plays a crucial\nrole in conjugating proteins\nor peptides under biological conditions. One widely used bioorthogonal\nreaction is strain-promoted alkyne–azide cycloaddition (SPAAC)\ndevoloped in the group of Carolyn R. Bertozzi.  Unlike traditional azide–alkyne cycloaddition, SPAAC\ndoes not require a copper catalyst. Instead of activating the alkyne\nwith Cu­(I), SPAAC introduces a strained cycloalkyne, such as difluorooctyne\n(DIFO), dibenzylcyclooctyne (DIBO), or biarylazacyclooctynone (BARAC). \n − \n \n  These strained cycloalkynes destabilize the alkyne, enhancing the\nreaction driving force and promoting the cycloalkyne to relieve its\nring strain.\nThe Staudinger reaction\ninvolves the reaction between a methyl\nester phosphine and an azide, leading to the formation of an aza-ylide\nintermediate, which is subsequently captured to yield a stable covalent\nbond. This cross-linking chemistry, initially developed in the early\n20th century by polymer chemist and Nobel Laureate Hermann Staudinger,\nhas recently gained prominence in biological systems as a bioconjugation\ntechnique. It exhibits essential characteristics for bioorthogonal\nchemistry, such as biocompatibility, selectivity, and rapid, high-yield\nturnover, making it applicable across a diverse range of applications.\nThis application in chemical biology is commonly referred to as Staudinger\nligation. \n ,\nThioether formation entails\nthe reaction between a thiol group\nand an electrophile (such as a haloalkane or sulfonate ester) to generate\na thioether linkage. A specific example of this process is thiol-maleimide\nconjugation, which occurs via a Michael addition mechanism between\nthiol (-SH) groups and maleimide to establish a stable thioether bond.\nThiol-maleimide conjugation is widely used to attach chemical labels\nto peptides and proteins, including fluorescent dyes, polyethylene\nglycol (PEG), radiolabels, antibodies, and small molecules.  The reaction offers several advantages, including\nrapid kinetics between maleimides and thiols, as well as a preference\nfor neutral pH conditions. However, it is not without challenges;\nside reactions, such as thiazine rearrangement, can occur during thiol-maleimide\nconjugation.  These side reactions are\noften attributed to the instability of the maleimide-cysteine conjugate.\nNotably, a significant increase in the rate of thiazine formation\nhas been observed at basic pH values, indicating a base-dependent\nmechanism that involves nucleophilic attack of the succinimide by\nthe N-terminal amine. Furthermore, substituting the amino acid adjacent\nto the N-terminal cysteine with various residues has resulted in the\ngeneration of thiazine impurities, albeit at different rates. Even\nwhen employing a maleimide linker designed for enhanced stability,\nconsiderable thiazine formation has been noted, suggesting the ubiquitous\nnature of this side reaction.  The presence\nof thiazine impurities has been confirmed using various analytical\ntechniques. Protonation of the N-terminal amino group in acidic conditions\ncan inhibit the nucleophilic reaction and subsequent thiazine formation.  However, performing conjugation under acidic\nconditions (around pH 5) necessitates subsequent purification and\ncareful handling of peptide conjugates to prevent the loss of succinimidyl\nthioether. An alternative approach to mitigate thiazine formation\ninvolves the acetylation of the N-terminal cysteine. Given the widespread\noccurrence of the thiazine side reaction, it is advisible to avoid\nusing N-terminal cysteine in peptide designs.\nThiol–ene\nclick chemistry encompasses the reaction between\na thiol group and an alkene group to form a thioether linkage.  This method offers several advantages, including\nhigh yields, stereoselectivity, rapid reaction rates, and favorable\nthermodynamic profiles. The addition reactions generally proceed through\ncatalyzed Michael additions or free-radical additions. In free-radical\nadditions, various stimulisuch as light, heat, or radical\ninitiatorscan be employed to generate thiyl radical species.\nThese radicals then react with the ene functional group via an anti-Markovnikov\naddition, resulting in the formation of a carbon-centered radical.\nFollowing this, a chain-transfer step occurs, wherein a hydrogen radical\nis removed from a thiol, allowing the process to continue through\nmultiple propagation steps.  This reaction\nis particularly valuable in radical-based photopolymerization because\nit can proceed quantitatively and rapidly through a straightforward\nmechanism under ambient atmospheric conditions. Depending on the thiol\nand ene functional groups involved, the carbon-centered radical is\ngenerated in this reaction.\nOther photochemical cross-linking\nstrategies involve the development\nof photoinducible reactions for conjugation, such as photoreactive\nsmall molecules like diazirines or benzophenones. Upon exposure to\nUV light, these molecules can be activated to form reactive intermediates\nthat cross-link nearby biomolecules, allowing for spatiotemporally\ncontrolled modification of peptides and proteins. Additionally, photoreactive\nmolecules can include amino acids, such as p-benzoylphenylalanine\n(pBPA), and photoreactive diazirine analogs of leucine and methionine. \n , \n  Upon exposure to ultraviolet light (UV), these molecules undergo\nactivation, enabling them to covalently cross-link proteins within\ntheir native protein–protein interaction domains  in\nvivo . This approach allows for the identification and characterization\nof both stable and transient protein interactions within cells, eliminating\nthe necessity for traditional chemical cross-linkers and solvents\nthat could interfere with the cellular biology under investigation\nin the experiment.\n\nSmall molecules often face limitations in modulating or interfering\nwith PPI. Their protein binding affinity tends to be lower than larger\nbiological modulators such as antibodies, proteins, and peptides.\nIn this context, short peptides and miniproteins emerge as promising\ncandidates for rationalizing peptide-based drugs, offering higher\ntarget affinity and potentially reduced toxicity compared to small\nmolecules.\nThe biological activity\nand function of peptides depend highly\non their ability to adopt specific shapes or conformations. Peptides\nwith rigid conformations exhibit reduced flexibility, which enhances\nselectivity, improves stability against protease degradation, and\nlowers toxicitykey attributes that make them strong candidates\nfor orally bioavailable peptide therapeutics. Synthetic approaches\ntoday aim to modify the topology of peptidomimetics, compelling them\nto assume specific conformations that stabilize the tertiary fold\nof proteomimetics or provide a more stable secondary structure under\ndiverse conditions ( Figure  \n  and  ).\nTypes of topologies that\ncan be achieved through peptide cyclization.\nA peptide ring can be formed via reactions between the N- and C-termini,\nbetween one terminus and a side chain, or between two side chains.\nIn all cases, the primary objective is to stabilize the peptide in\na specific conformation or to restrict the number of possible conformations,\nthereby reducing its flexibility. Created in BioRender.\nCommon chemical strategies for peptide cyclization include\ndisulfide\nbridge formation, lactamization, click reactions such as CuAAC, and\nmetal-mediated cyclization where the metal is chelated by side chains\nor termini. Multiple cycles can also be introduced, as in the formation\nof a bicyclic structure using a scaffold, e.g., Chemical Linkage of\nPeptides onto Scaffolds (CLIPS). Cyclization methods also encompass\nstapling and stitching, which often involve click reactions or the\nformation of a carbon–carbon bridge. Created in BioRender.\nIn natural proteins, topology and flexibility are\noften altered\nby intramolecular cross-links, such as disulfide bridges strategically\nplaced between distinct secondary structures to stabilize the tertiary\nfold. Other methods utilized other natural amino acids to form metal-mediated\nbonds, or lactam groups. \n − \n \n  It is important to note that\nthe same approaches can be employed to stabilize structures beyond\nhelices or produce macrocyclic structures. For instance, naturally\noccurring Cys-Cys bonds are found in β-sheets (e.g., defensins)\nand are harnessed to stabilize β-hairpins and dimers of β-sheets.  Similarly, click chemistry stapling techniques\ncan confer stability to β-turns and β-hairpins.  Macrocyclic peptides have demonstrated suitability\nas pharmaceuticals and recent advances have moved beyond mimicking\nnatural cyclic peptides.  Medium-size\npeptides are emerging as molecules bridging the gap between small\nmolecules and biologics, showing potential to target previously challenging\nproteins to interact with using small molecules.\nIn addition to the macrocyclization mentioned for\nstabilizing peptide\nstructures, lasso peptides represent another fascinating class of\nnaturally occurring, highly stable, and structurally unique peptides.\nSynthesized ribosomally by microorganisms, these peptides feature\na knot-like structure in which the peptide backbone forms a loop covalently\nthreaded through an amino acid side chain, resulting in a “lasso”\nshape.  This configuration imparts remarkable\nstability, protecting the peptide from degradation and enhancing its\nbioactivity. Lasso peptides have demonstrated a wide range of biological\nactivities, particularly in antimicrobial defense. Their rigidity,\ndue to the knot formation, makes them ideal candidates for pharmaceutical\ndevelopment. Indeed, lasso peptides are increasingly being engineered\nfor therapeutic use, showing promise as antimicrobials or antitumor\nagents. \n , \n  Their ability to bind tightly and selectively\nto target proteins or enzymes opens new avenues for drug discovery,\nparticularly for challenging biological pathways that are difficult\nto target with conventional small molecules. For readers interested\nin learning more about lasso peptides, we encourage starting with\nexcellent papers authored by the research groups led by Mitchell,\nMarahiel, and Swanson. \n −\nWithin the field of peptidomimetics, disulfide bonds serve as constrained\nstructural elements frequently employed to generate macrocycles. This\npractice helps immobilize the peptide in its bioactive conformation,\nthereby enhancing the pharmacological properties of peptides. Despite\ntheir predominant role in these applications, disulfide bonds exhibit\nmultifaceted functions, participating in oxidative folding and other\nbiological processes.  Due to their involvement\nin redox reactions and biological processes, disulfide bridges are\nnot an ideal choice for generating stable cyclic peptides, and alternative\nstrategies have therefore been explored.\nOne approach to mimicking\nor replacing disulfide bonds is the use of bis-electrophilic linkers.\nThese linkers come in various forms, including those based on alkylation,\nacylation, Michael addition, nucleophilic aromatic substitution, and\nmetal-mediated couplingall of which exploit the unique nucleophilicity\nof sulfur.  Linkers with two identical\nelectrophilic groups are primarily limited to intramolecular processes\nsuch as macrocyclization, stapling, and disulfide rebridging, or to\nthe formation of homodimers ( Figure  \n a-c). To enable selective cross-conjugation, nonsymmetrical\nlinkers with sufficiently different reaction rates between their electrophilic\ngroups are required. Maleimide-succinimidyl esters ( Figure  \n d and e) are among the most\nwidely used heterobifunctional cross-linkers in bioconjugation. However,\nmaleimide conjugates can present stability issues in biological systems,\nand the activated ester in ( Figure  \n d) is rapidly hydrolyzed in alkaline aqueous media,\nreacting with both thiols and amines. This issue can be mitigated\nby replacing the ester with an azide ( Figure  \n e) to enable bioorthogonal reactions, but\nthis modification prevents the use of natural amino acids as conjugation\npartners.\nBis-electrophilic linkers for bioconjugation and mimicking disulfide\nbonds. Top row: homobifunctional linkers. Bottom row: heterobifunctional\nlinkers.\nA related study explores the use of 1,4-dinitroimidazoles\nfor macrocycle\nformation. These compounds function as highly efficient bifunctional\nbioconjugation reagents, reacting with cysteine side chains under\naqueous acidic and neutral conditions via a  cine -substitution\nmechanism to form stable products.  In\nthese conditions, 1,4-dinitroimidazoles react selectively with cysteine.\nHowever, in organic solvent and with base, 1,4-dinitroimidazoles can\nalso react with lysine through a ring-opening and ring-closing mechanism\n( Scheme  \n ). By exploiting\ntheir ability to react with both cysteine and lysine via distinct\nmechanisms, these reagents enable the formation of bioconjugates with\nsuperior chemoselectivity and stability compared to conventional maleimide–thiol\nconjugates.\nThe research groups of Wade and Hossain developed\nstrategies to\nreplace disulfide bonds in insulin. Mature insulin is stabilized by\nthree disulfide bonds: two linking the A and B chains and one within\nthe A chain. Although the synthesis of the two chains was achieved\nmany years ago, correctly forming the disulfide bonds to link and\nstabilize them remains a significant challenge. Many approaches rely\non orthogonal cysteine protection and regioselective disulfide bond\nformation, while others focus on directly mimicking the disulfide\nbridge. The groups employed cystathionine to replace the A6-A11 intrachain\ndisulfide bond, leading to enhanced thermal stability.  This approach involved substituting a disulfide\nbridge with a thioether linkage, with cystathionine being generated\nin situ from cysteine using orthogonal protection strategies. Further\ndetails on the mechanism are illustrated in  Scheme  \n .\na This strategy begins\nby substituting\na disulfide bond with a thioether linkage, followed by peptide elongation\nthrough solid-phase synthesis. The cystathionine bridge is generated\nin situ from cysteine residues using orthogonal protection strategies.\nAfter cystathionine formation, peptide chain elongation proceeds to\ncomplete the full sequence (not shown). The scheme also details the\nreagents and conditions used up to the formation of the cystathionine\nlinkage.\nThis strategy begins\nby substituting\na disulfide bond with a thioether linkage, followed by peptide elongation\nthrough solid-phase synthesis. The cystathionine bridge is generated\nin situ from cysteine residues using orthogonal protection strategies.\nAfter cystathionine formation, peptide chain elongation proceeds to\ncomplete the full sequence (not shown). The scheme also details the\nreagents and conditions used up to the formation of the cystathionine\nlinkage.\nHirudin is a 65-amino acid peptide\nthat contains three disulfide\nbridges. In a study, selenium was used as a substitute for sulfur\nin cysteine to investigate the effects of diselenide bridges on folding,\nstructure, and activity, both at native and non-native positions.  Three designed analogues incorporated diselenide\nbonds at the native cross-links (6–14, 16–28, and 22–39),\nwhile a fourth analogue introduced a diselenide bridge at a non-native\nposition (6–16), based on the proposed role of this non-native\ndisulfide bond in the early stages of hirudin folding. Overall, the\nresults indicate that replacing native disulfide bonds with diselenide\ncross-links enhances folding efficiency toward the native state, significantly\nreducing the formation of nonproductive intermediates. Notably, even\nthe non-native diselenide-containing analogue (6–16) exhibited\na similar improvement in folding efficiency.\nMacrolactamization\nis a widely utilized strategy to build a bridge\ndue to several key advantages, including its mild reaction conditions,\nwhich are compatible with sensitive functional groups, high chemoselectivity\nfor desired cyclic structures, and versatility stemming from the broad\navailability of starting materials such as amines and carbonyl-containing\ncompounds. Moreover, its biological relevance is underscored by the\nprevalence of lactam rings in natural products and bioactive molecules.\nDespite these merits, macrolactamization is limited by challenges\nsuch as steric hindrance, substrate incompatibility due to reactivity\nissues, and potential polymerization side reactions caused by competition\nbetween inter- and intramolecular processes. The incorporation of\nturn-inducing elements (TIEs), such as proline, has been shown to\neffectively address these challenges, facilitating the synthesis of\na miniprotein with a small β-sheet structure.  A critical loop connects the two β-strands, promoting\nprotein–protein interactions (PPI). This design draws inspiration\nfrom the VP3VR-VIII region of the adeno-associated virus (AAV) capsid\nprotein.\nAdditionally, strategies involving hydrocarbon bridges\nto “staple”\npeptides across side chains or hydrogen bond surrogates in the backbone\nhave proven effective in producing biologically functional molecules\nstabilizing the helical structure.  Stapling\ntechniques employing non-natural elements enhance protease resistance\nand potency both in vitro and in vivo.\nAnother general cyclization\nmethod has been developed, drawing\ninspiration from nonribosomal peptide synthetases (NRPSs).  Natural cyclic peptides, such as rufomycin\nand cyclosporin A, are biosynthesized by NRPSs, which possess the\nability to incorporate unnatural amino acids and introduce diverse\nmodifications, such as  N -methylation, epimerization,\nand oxidation, during peptide synthesis. The total synthesis of nonribosomal\ncyclic peptides (NRcPs) traditionally relies on standard coupling\nreagents. However, this approach is often labor-intensive, requiring\nextensive use of protecting groups, prolonged reaction times, and\nyielding side reactions like epimerization and dimerization, which\nlower the overall efficiency. The newly developed method, inspired\nby the biosynthetic cyclization processes of NRPSs, enables the production\nof macrocycles with remarkable speed (within minutes), high selectivity,\nand excellent yield. This general approach to NRcP synthesis and macrocyclization\nis effective regardless of sequence or ring size. The process involves\nthe synthesis of a linear peptide hydrazide via standard solid-phase\npeptide synthesis. After complete deprotection, the hydrazide is oxidized\nto an azide, facilitating tail-to-head cyclization. The study also\nhighlights the critical role of pH and solvent choice. Oxidation of\nthe hydrazide is most efficient in water under acidic conditions,\nwhereas the actual cyclization step proceeds optimally in an organic\nsolvent at neutral pH. This biphasic method achieves rapid cyclization\nin just a few minutes, offering a highly efficient and versatile strategy\nfor cyclic peptide synthesis.\nNatural cyclic peptides, exemplified by valinomycin and various potent\nionophores with metal-mediated side-chain links within their structures,\nhave inspired scientists to leverage metals to enclose macrocycle\npeptides. Metals play a dual role, not only facilitating cyclization\nbut also influencing the secondary structure of peptides.\nThe\ninitial applications of metal-mediated cyclization involved the dimerization\nof peptide methyl esters and the binding of carbonyl and amide groups\nat peptide termini.\nMetal–ligand\ninteractions play a pivotal role in directing\npeptide structural control by stabilizing helices and facilitating\nthe formation of coiled coils and multihelical complexes. Peptides\nthat are shorter than 15 residues typically struggle to adopt α-helical\nstructures. However, the introduction of metal ions can assist in\nhelix formation by creating bridges that enhance structural stability.\nThis metal-assisted stabilization technique finds applications in\ninvestigating protein folding, designing peptidomimetics, and developing\ninhibitors.\nTransition metals, notably Ni 2+ , Zn 2+ , Cd 2+ , and Cu 2+ , are frequently employed\ndue to their\nability to form stable complexes with side chains of histidines, cysteines,\nor non-natural amino acids featuring two carboxylic groups. The formation\nof these complexes leads to the generation of macrocycles, effectively\nstabilizing the peptide backbone.  It\nis of significance to note that alkali metals (such as Li + , Na + , and K + ) lack the ability to chelate\nside chains on peptides. However, they exhibit the capability to transform\nrandom coils into rigid helices. This transformation is particularly\nsignificant for coiled-coil structures, which are assemblies formed\nby helical sequences characterized by amino acid positions typically\ndenoted as a-g. \n , \n  Coiled coils are formed by hydrophobic\ninteractions, primarily involving residues located at the a and d\npositions. Additionally, the e and g positions play a crucial role\nin stabilizing these structures through ionic interactions, specifically\nwhere the g position of one helix interacts with the e position of\nanother. Metal ions can be strategically employed to facilitate a\ncontrolled folding transition of coiled coils. By incorporating high-affinity\nmetal binding sites at the e and g positions, one can effectively\nstabilize the coiled-coil formation. When both binding sites contain\nnegatively charged side chains, they repel each other in the absence\nof the preferred metal ion, thus preventing assembly. The introduction\nof metal ions mitigates this repulsion and promotes the correct alignment\nand stabilization of the coiled-coil structure. Moreover, zinc ions\nhave been utilized to mediate bridging and create a 16-helix arrangement\nwith four copies of cytochrome cb562 (cyt cb562). Each cyt cb562 represents\na 4-helix bundle heme-containing protein. In this context, Zn­(II)\ncoordinates the di-His motifs on the surface of cyt cb562 (PDB:  2QLA ). The coordination\nof Zn-His plays a crucial role in protein multimerization, as evidenced\nby the dissolution of aggregates upon adjusting the pH below 6 and\ntreating with EDTA.\nSeveral studies\nhave employed metals to facilitate peptide cyclization.\nAn early example is the use of silver ions, which enable the cyclization\nof unprotected or minimally protected peptides. All Ag + -assisted cyclizations of minimally protected peptides were conducted\nin aqueous acetate-buffered solutions at pH 5–6, for two main\nreasons.  First, the affinity of the\nAg +  ion follows the order S ≫  N  > O. Thus, coordination of one or more Ag +  ions between\nthe nitrogen atom of the N-terminal amino group and the sulfur atom\nof the C-terminal thioester could promote the formation of the desired\ncyclic intermediate. Second, under these mildly acidic aqueous conditions,\nAg + -mediated hydrolysis proceeds slowly while selectivity\nfor aminolysis remains high. This reaction exhibits chemoselectivity\ntoward the formation of lactams and lactones, typically requiring\ntwo equivalents or more of silver ions and a reaction time of approximately\n2 h to achieve complete cyclization.\nAnother strategy involves\nexploiting the strong affinity between\nnickel ions and histidine residues. Positioning three histidine residues\nat each terminus of a peptide with low intrinsic propensity for independent\nsecondary structure formation can induce an ordered conformation in\nthe presence of nickel. Structural studies conducted via NMR further\ndemonstrate that the insertion of a single histidine residue at each\nend of short bioactive peptides promotes a more compact and predictable\nfolding pattern, without significantly altering the peptide backbone.\nThe metal ion in these reactions is not\nmerely part of the cyclization\nprocess but also acts as a type of catalyst. For example, natural\npeptides composed of five or seven residues have been synthesized\nand cyclized using 3-(diethoxyphosphoryloxy)-1,2,3-benzotriazin-4­(3H)-one\n(DEPBT) as a coupling reagent in solution, with the process mediated\nby different metal ions.  Although the\nlinear peptides lack side chains capable of strong metal complexation,\nmetal ions such as Fe 2+ , Ni 2+ , Zn 2+ , and Cr 3+  were found to strongly coordinate with the\ncarboxyl groups. In contrast, alkali metal ions such as Li + , Na + , K + , Rb + , and Cs +  do not form strong complexes like transition metals, but they can\ncoordinate to the oxygen atoms of carbonyl and amide groups near the\nC-terminus with low affinity. This coordination promotes the formation\nof a turn structure, as demonstrated by CD spectroscopy studies. The\nresulting turn brings the N- and C-termini of the linear peptide into\nproximity, thereby enhancing the efficiency and yield of cyclization.\nThe groups of Pentelute and Buchwald have reported that palladium­(II)\ncomplexes can be employed for efficient and highly selective cysteine\nconjugation reactions, which proceed rapidly and under a broad range\nof biocompatible conditions.  The straightforward\nsynthesis of these palladium reagents from a variety of readily available\naryl halides and trifluoromethanesulfonate precursors makes the method\nhighly practical, enabling access to a wide structural space for peptide\nand protein modifications. Palladium reagents bearing two electrophilic\nmetal centers were effectively utilized to cross-link two cysteine\nresidues within a peptide chain, thus allowing the generation of stapled\npeptides featuring various aryl linkers. Notably, performing the reaction\nat a peptide concentration of 10 mM in a 1:1 (v/v) acetonitrile/water\nmixture at pH 7.5, with a 2-fold excess of the bis-palladium complex\n2A, led to the quantitative formation of the desired stapled peptide\nwithin 10 min. The resulting aryl bioconjugates demonstrated high\nstability against acids, bases, oxidants, and external thiol nucleophiles.\nThese palladium complexes show considerable promise as practical benchtop\nreagents for diverse bioconjugation applications.\nSimilarly,\nthe same groups demonstrated that, in the presence of\na biarylphosphine-supported palladium­(II)–aryl complex and\na weak base (sodium phenoxide, p K \n a  = 10),\nlysine amino groups in unprotected peptides underwent C–N bond\nformation at room temperature.  This\nreaction and the developed protocol enable the formation of N–aryl\nconjugates, which exhibit greater stability compared to their corresponding\nS–aryl counterparts. This approach proved effective for the\nconjugation of a variety of organic compounds, including peptides,\nwhich were successfully cyclized.\nMacrocycles can\nbe formed by creating a bridge between residues aligned on the same\nface of the helix, typically at positions i, i+4, i+7, and i+11, with\ni+4 and i+7 being the most common. Early stapling strategies involve\nusing natural amino acids for side chain-to-side chain cross-linking.\nExamples include the use of lactam between Lys and Glu/Asp residues,\nthioether between two Cys residues, His-His via metal chelates, and\nvarious other methods involving proteinogenic amino acids and synthetic\napproaches.  Utilizing natural amino\nacids as anchoring points necessitates either selective orthogonal\nprotection or the replacement of identical amino acids with different\nones within the peptide sequence, imposing limitations on this peptide\nstapling approach.\nOne of the well-established peptide stapling\nstrategies that overcome these limitations involves the use of unnatural\namino acids, specifically through hydrocarbon stapling. This technique\nemploys Grubbs catalysts to link the side chains of two non-natural\namino acids in the solid phase at positions i, i+4, or i, i+7. The\nGrubbs catalyst, [(PCy 3 ) 2 Cl 2 Ru =\nCHPh], plays a pivotal role in initiating the formation of a carbon–carbon\nbridge, connecting specific α,α-disubstituted amino acids\nwith olefinic side chains through a ring-closing metathesis (RCM)\nreaction.  The hydrocarbon bridge connects\nat two locations along a synthetic peptide backbone and stabilizes\nthe α-helical arrangement forming a macrocycle with increased\nstability and hydrophobicity.\nStapling also emerges as a method\nto constrain and stabilize non-natural\npeptide foldamers into helical-mimicking conformations. In an initial\nstudy, the Hoveyda-Grubbs generation II catalyst was employed to staple\nβ-peptides.  These stapled peptides\nexhibited helicity in a pure phosphate buffer and various solvents,\nincluding TFE, methanol, and a combination of acetonitrile and buffer.\nAn extensively utilized strategy for chemical ligation and peptide\nstapling involves the Cu­(I)-catalyzed azide–alkyne 1,3-dipolar\nHuisgen cycloaddition, commonly known as the CuAAC click reaction\nor the strain-promoted azide–alkyne cycloaddition (SPAAC) (see  Section  \n ).\nIn general, stapling methods involving\nnatural amino acids, Grubbs\ncatalysts, or click chemistry constitute one-component stapling techniques,\nallowing the direct coupling of complementary side-chain groups. In\ncontrast, two-component stapling employs a bifunctional linker compound\nthat reacts with two complementary non-native amino acids in the peptide\nto form a staple.  This stapling technique\ninvolves reacting linear i,i+7 diazido peptides (i.e., containing\ntwo azido amino acids that are seven residues apart) with dialkynyl\nstapling linkers under Cu­(I) catalysis. As this reaction produces\npeptides bearing a bis-triazole linkage, this process is called double-click\nor two-component stapling.  The primary\nadvantage of two-component stapling lies in the ability to introduce\nmore diverse staple linkages without the need for synthesizing complex\nunnatural amino acids. However, the more intricate reaction pathway\nin two-component stapling may lead to the generation of more byproducts\ncompared to one-component stapling. One example of a competing path\ninvolves coupling two linker moieties to a single peptide, one at\neach non-native amino acid.\nMost two-component stapling techniques\nare adaptations of their\none-component stapling counterparts, with bis-lactamization being\nan example.  The use of natural amino\nacids simplifies the synthesis of linear peptides due to their cost-effectiveness\nand availability, minimizing alterations to the wild-type peptide\nsequence and potentially avoiding negative impacts on binding affinity.\nHowever, challenges may arise regarding orthogonality and chemoselectivity.\nMany two-component stapling strategies that utilize natural amino\nacids primarily focus on lysine and cysteine, with limited applicability\nto tryptophan. In contrast, employing unnatural amino acids for peptide\nstapling requires either procuring or synthesizing these nonproteinogenic\namino acids, which can be both costly and time-consuming. Nonetheless,\nthis approach offers excellent orthogonality and allows for a variety\nof staple compositions.\nAnother\nstapling method was used by the Pentelute group who identified\nan efficient transformation process involving perfluoroaromatic molecules\nand a cysteine thiolate, leading to arylation at room temperature.  This method allows for the selective modification\nof cysteine residues in unprotected peptides, enabling the incorporation\nof rigid perfluoroaromatic staples. When applied to a peptide sequence\ndesigned to interact with the C-terminal domain of the HIV-1 capsid\nassembly polyprotein (C-CA), this stapling modification resulted in\nimproved binding affinity, cell permeability, and proteolytic stability\ncompared to its unstapled counterpart. Importantly, the chemical stability\nof the resulting staples facilitated their use in the native chemical\nligation-mediated synthesis of a small protein capable of binding\nto the human epidermal growth factor receptor 2 (HER2). The same research\ngroup reported a mild and efficient method for synthesizing macrocyclic\npeptides via nitrogen arylation from unprotected precursors. They\nexplored various electrophiles and lysine-based nucleophiles, successfully\ngenerating high-yield products in a macrocyclization scan that included\n14 different variants. The nitrogen-linked aryl products demonstrated\ngreater stability against base and oxidation than thiol-arylated counterparts,\nhighlighting the advantages of this methodology.  Notably, when this N-aryl macrocyclization was applied\nto a p53 peptide inhibitor of MDM2, it led to the discovery of a nanomolar\nbinder with improved proteolytic stability and cell permeability.\nPeptides that contain two macrocyclic structures are commonly referred\nto as bicyclic peptides. This classification includes peptides with\ntwo loops formed by a scaffold anchored at three points within the\npeptide sequence. Additionally, macrocyclic peptides featuring an\ninternal bridge are also categorized as bicyclic peptides, a structural\nmotif frequently observed in nature. Some definitions may also encompass\npeptides with double-stapled or double-macrocyclic configurations.\nResearch efforts continue to explore various bicyclic topologies,\naiming to establish novel synthetic pathways. \n −\nBicyclic peptides have gained prominence as a significant\nsubset within the constrained peptide family and are expected to possess\nsubstantial therapeutic potential, as evidenced by the growing interest\nreflected in scientific literature. \n , − \n \n  Compared to monocyclic peptides, bicyclization offers several advantages,\nincluding enhanced structural rigidity, improved metabolic stability,\ncell permeability and increased target affinity.  Their binding characteristics, similar to those of antibodies,\nenable them to effectively disrupt protein–protein interactions.\nWith two macrocyclic structures, these peptides can engage with a\nsingle target structure or simultaneously bind to two different targets.\nAs a result, bicyclic peptides are frequently designed for applications\nin antimicrobial or anticancer therapies, making them a compelling\narea of research for next-generation pharmaceuticals. \n , \n  This growing interest is underscored by the emergence of several\ncompanies, such as Bicycle Therapeutics and Pepscan, that focus primarily\non developing bicyclic peptides.\nChemical Linkage of Peptides\nonto Scaffolds (CLIPS) is the methodology\nto produce bicycles wherein peptides are cyclized and tethered onto\na central scaffold to impart rigidity and stability. The central scaffold\nmay consist of organic molecules, dendrimers, or other polymeric structures.\nBy attaching peptides at multiple points to the scaffold, a precisely\ndefined three-dimensional structure is created. The benefits of CLIPS\ninclude precise control over peptide conformation and spatial arrangement,\nthe ability to present multiple peptides in a defined orientation,\nand enhanced binding properties due to the stable and rigid conformation.\nCLIPS finds application in the development of vaccines and immunogens\nby presenting epitopes in a native-like conformation.\n1,3,5-tris­(bromomethyl)­benzene (TBMB)\nis widely utilized as a reagent\nfor synthesizing bicyclic peptides through cysteine alkylation. Its\napplication in conjunction with phage-displayed proteins has represented\na significant stride in the development of genetically encoded bicyclic\npeptide libraries. Integration of TBMB with the phage display platform\nenables the rapid identification of bioactive bicyclic peptides through\niterative selections, presenting a molecularly lighter alternative\nto antibodies and other binding proteins. \n , \n  However, excessive TBMB usage may induce nonspecific modification\nof linear peptides. To enhance this technique, additional reagents\nwith similar symmetry and thiol-reactivity, 1,3,5-triacryloyl-1,3,5-triazinane\n(TATA), 1,3,5-tris­(bromomethyl)­benzene (TBMB), and N,N’,N’’-(benzene-1,3,5-triyl)­tris­(2-bromoacetamide)\n(TBAB), have been developed ( Figure  \n ). \n ,\nThree common organic\nlinkers applied for the cyclization of bicyclic\npeptides.\nIn recent years, innovative approaches have been\ndeveloped for\nsynthesizing bicyclic peptides utilizing the triple cysteine motif.\nMuch like TBMB alkylation, these strategies often capitalize on the\nunique nucleophilicity of cysteine residues. Pentelute and colleagues\npreviously established a peptide stapling method using perfluoroaryl\nlinkers.  To adapt this technique for\nconnecting three cysteine residues, they initially employed an excess\nof decafluorobiphenyl (DFBP) to monosubstitute each cysteine side\nchain. Following this step, benzene-1,3,5-trithiol (BTT) was introduced\nto the modified peptide, enabling bicyclization through a second nucleophilic\naromatic substitution.  Double-stapled\npeptides can be synthesized from four cysteines if two are orthogonally\nprotected with StBu.\nTo expand beyond cysteine modifications,\nChen and colleagues developed\na reaction that utilizes lysine and arginine residues for cyclization.  This method, based on a stapling strategy that\nemploys formaldehyde to link amino acids, allows for the creation\nof multicyclic peptide topologies. By introducing formaldehyde along\nwith amine and guanidine, they successfully achieved the cyclization\nof peptides containing two lysines and one arginine residue. Alternatively,\nby substituting the arginine side chain with a lysine residue and\nusing guanidine as a reagent, a connection between three lysines could\nbe formed. However, the inclusion of basic amino acids in this bicyclization\nlimits the sequence diversity of potential bicyclic peptides that\ncan be constructed using canonical amino acids.\nAs in macrocyclization,\nmetals can also assist in the formation\nof bicyclic structures. For instance, in the case of a triple-cysteine\npeptide, Stauber et al. used Au 3+  and complexes such as\nthe  tert -butyl substituted aminophosphine-supported\nAu­(III) complex, known as the (P,N) supported Au­(III) complex.  The trimetallic (P,N) supported Au­(III) complex,\ndenoted [3] 3+ , which features three metasubstituted (P,N)­Au­(C 6 H 4 )Cl fragments surrounding a central aryl anchor,\nwas synthesized. Treatment of 1,3,5-tris­(4-iodophenyl)­benzene with\nthe (P,N) supported Au­(III) complex in the presence of AgSbF 6  resulted in clean conversion to [3] 3+  at room temperature.\nEfficient cyclization of the model linear tricysteine peptide, H 2 N–GCAENCAFGCA–CONH 2 , via its three\ncysteine thiols was achieved by treating the peptide with complex\n[3] 3+  in a TRIS buffer and MeCN solvent mixture.\nBi­(III) was introduced to overcome the limitations associated with\nTATA, TBAB, and TBMB. While TBMB binds irreversibly, potentially modifying\nother reactive peptide residues even when used in slight excess, scaffolds\nsuch as TATA and TBAB were specifically developed to stabilize peptide\nconformations by promoting hydrogen bond networks. However, all conventional\nscaffolds contain flexible bonds, which ultimately limit their ability\nto fully rigidify peptide bicycles. Bi­(III) is nontoxic, selective,\nstable, and rigid, and it effectively interacts with cysteine residues\nin peptides and proteins. Peptide–bismuth bicycles form instantaneously\nat physiological pH, are stable in aqueous solutions for extended\nperiods, and exhibit significantly higher resistance to proteolysis\ncompared to their linear precursors.  These bicyclic peptides show up to 130 times greater activity and\n19 times more proteolytic stability than their linear analogs without\nbismuth. Additionally, they target proteases from Zika and West Nile\nviruses, unveiling a new lead compound with inhibition constants of\n23 and 150 nM, respectively.\nMetal-mediated S-arylation approaches\nrequire only one reaction\nto bicyclize a.[ \n ,  \n ] Stauber\net al. developed a number of Au­(III)-complexes not only limited to\nbicyclization, but also suitable for mediating multisite bioconjugation\nand peptide stapling.[ ] Bicyclic peptides were generated in a mixture of neutral buffer\nand acetonitrile, whereby the central scaffold of various Au­(III)\ncomplexes was transferred onto the three cysteines of a linear peptide\n( Scheme  \n ).[ ] Mudd et al. later expanded\nthis work by introducing further Au­(III) complexes that contained\nsmaller scaffolds ( Scheme  \n ).[ ] Previously,\nChen and co-workers had found an alternative pathway to effectively\ncreate similar bicycles using Pd-catalyzed S-arylation with triiodoarenes\n( Scheme  \n ,  Figure  \n b).[ ]\nRecent strides\nhave been taken in designing intramolecular bicyclization\nreactions, departing from traditional methodologies employing external\nreagents on assembled linear peptides. This innovative approach involves\nintegrating a reactive handle into the linear peptide chain. Reymond\nand collaborators exemplified this by coupling 3,5-bis­(chloromethyl)-4-methylbenzoic\nacid to the peptide N-terminus, leading to the subsequent formation\nof two thioether linkages with cysteines.  Through this strategy, they synthesized bicyclic antimicrobial peptides\neffective against multidrug-resistant strains of  Acinetobacter\nbaumannii  and  Pseudomonas aeruginosa . Another\nstrategy entails synthesizing an amino acid with two carboxylic acids\nduring solid-phase peptide synthesis. A photoreaction was employed\nto introduce two 3-mercaptopropionic acid molecules to propargylglycine,\nenabling selective internal amide couplings following orthogonal deprotection\nof two amines. Notably, a dual-targeted, α-helical bicycle synthesized\nvia this method exhibited potential as a cytotoxin for cancer treatment.\nNatural bicyclic peptides frequently feature internal cross-links,\nmaking this topology a significant target for chemical synthesis.\nRecent examples include various bridges within macrocyclic peptides,\nsuch as FF, FY, and YY-like biaryl linkages formed in cyclic peptides.\nThe synthesis of these structures generally requires microwave-assisted\nSuzuki-Miyaura cross-coupling conditions. Teixidó and collaborators\nutilized this cross-coupling technique to connect two tryptophan residues\nin cyclic peptides, enabling homocouplings at different positions\nusing various bromotryptophan derivatives.\nBicyclic peptides exhibit increased stability and improved\ncell\npermeability, making them promising candidates for drug development.\nThe Grossman group illustrated that the formation of the bicyclic\nstructure can enhance the β-sheet character of the macrocycle,\nthereby improving its ability to penetrate cells. Their research focused\non identifying a novel target for β-catenin, which is a central\nhub for intracellular interactions within the Wnt signaling pathway.\nThey reported the creation of a library of β-sheet-mimicking\nbicyclic peptides that specifically target β-catenin, compete\nwith transcription factors for binding, and inhibit Wnt signaling\nin cellular contexts.\n\nAdvancing synthetic chemistry through the\nintegration of biotechnology\ninvolves harnessing the power of biological systems and techniques\nto enhance traditional chemical synthesis methods. This interdisciplinary\napproach combines principles from chemistry, biology, and engineering\nto develop innovative strategies for creating complex molecules with\nimproved efficiency, selectivity, and sustainability. By leveraging\nthe capabilities of biological systems, such as enzymes, microorganisms,\nand genetic engineering tools, researchers can overcome challenges\nin traditional synthetic chemistry and unlock new opportunities for\ndrug discovery, and chemical manufacturing.\nOne powerful technique\nwithin this integration is directed evolution,\na method that exemplifies how biotechnology can be used to accelerate\nthe development of novel chemical compounds. Directed evolution enables\nthe generation of diverse peptide sequences with desired properties\nthrough iterative rounds of mutagenesis, selection, and amplification. \n , \n  This approach starts with a known peptide sequence or scaffold,\nwhich is subjected to random or targeted mutations using techniques\nsuch as error-prone PCR or DNA shuffling. The resulting library of\npeptide variants is screened or selected for specific activities or\nproperties of interest. Selected peptides are then subjected to further\nrounds of mutagenesis and selection to optimize their performance.  Direct evolution allows for the creation of\npeptide sequences that may not exist in nature, providing access to\na vast sequence space beyond what is found in biological sources.\nLibraries derived from biological sources are often synthesized\nand screened using bacteriophages (phage display) or cell-free technologies\nsuch as mRNA display ( Figure  \n ).\nDrug discovery using display techniques. (a) In phage\ndisplay,\npeptides or proteins for screening are expressed and displayed on\nthe surface of bacteriophages. After interaction with the target,\npeptides or proteins with low affinity are washed away, while those\nwith higher affinity are retained and subjected to further rounds\nof screening. Iterative cycles of screening progressively enrich for\ndrugs with higher target affinity. (b) In mRNA display, a similar\niterative screening process is employed, but potential drug candidates\nare covalently linked to the mRNA from which they were synthesized.\nCreated in BioRender.\nPhage display is a powerful technique utilized\nto investigate interactions\namong proteins, peptides, and DNA This method leverages bacteriophagesviruses\nthat specifically infect bacteriato associate proteins with\ntheir corresponding genetic sequences.  In phage display, a gene encoding the protein of interest is inserted\ninto a gene responsible for a phage coat protein, resulting in the\nphage displaying the protein on its surface while the genetic information\nis contained within.  This arrangement\ncreates a direct linkage between the genotype and phenotype. The displayed\nproteins can then be screened for interactions with other proteins,\npeptides, or DNA sequences, facilitating the identification of binding\npartners. As a result, extensive libraries of proteins can be screened\nand selectively amplified through a process called in vitro selection,\nwhich emulates the principles of natural selection.\nmRNA display\nis an innovative technique utilized for the in vitro\nselection and evolution of proteins and peptides, enabling the generation\nof molecules with high affinity for specific targets. \n , \n  This process involves the creation of translated peptides or proteins\nthat are linked to their corresponding mRNA progenitors through a\npuromycin linkage. During the selection phase, these fusion molecules\ninteract with an immobilized target via affinity chromatography. Molecules\nexhibiting strong binding affinities are then reverse transcribed\ninto complementary DNA (cDNA), followed by amplification of their\nsequences using polymerase chain reaction (PCR). This results in the\ngeneration of a nucleotide sequence that encodes a peptide with a\nhigh affinity for the target. Puromycin functions as an analogue of\nthe 3′ end of tyrosyl-tRNA, mimicking both adenosine and tyrosine.\nIn mRNA display, all mRNA templates have puromycin linked to their\n3′ ends. As translation occurs, the ribosome traverses the\nmRNA template, and upon reaching the 3′ end, the attached puromycin\nenters the ribosome’s A site and is incorporated into the growing\npeptide chain. This incorporation leads to the release of the mRNA-polypeptide\nfusion from the ribosome. Unlike the cleavable ester bond found in\ntyrosyl-tRNA, puromycin possesses a nonhydrolyzable amide bond, which\ndisrupts translation and causes the premature release of the translation\nproducts.\nNot only linear peptides\nbut macrocyclic peptides can be produced\nthrough biological methods, with libraries generated using various\ntechniques: disulfide bridge formation: cysteine residues can be\nrandomly incorporated into sequences displayed on the surface of a\nphage, allowing the formation of disulfide bridges. A notable variant\nof this approach is the phage display combined with CLIPS cyclization\ntechnology. head-to-tail cyclization:\nthis method leverages the\nprotein splicing capability of split inteins to achieve intracellular\ncyclization, a technique known as SICLOPPS (Split Intein Mediated\nCircular Ligation of Peptides and Proteins). in vitro cyclization: linear peptide libraries encoded\nby mRNA are translated in vitro and subsequently cyclized using chemical\nreagents. An example is represented by the RaPID technology.\ndisulfide bridge formation: cysteine residues can be\nrandomly incorporated into sequences displayed on the surface of a\nphage, allowing the formation of disulfide bridges. A notable variant\nof this approach is the phage display combined with CLIPS cyclization\ntechnology.\nhead-to-tail cyclization:\nthis method leverages the\nprotein splicing capability of split inteins to achieve intracellular\ncyclization, a technique known as SICLOPPS (Split Intein Mediated\nCircular Ligation of Peptides and Proteins).\nin vitro cyclization: linear peptide libraries encoded\nby mRNA are translated in vitro and subsequently cyclized using chemical\nreagents. An example is represented by the RaPID technology.\nSICLOPPS, or\nsplit-intein circular ligation of peptides and proteins, is a method\nfor synthesizing cyclic peptides within cellular environments. It\noffers a robust approach with high efficiency and purity. This technique\ncan generate libraries containing up to 10 8  cyclic peptides.  It operates based on protein splicing, a natural\nprocess involving the removal of an internal protein segment, known\nas an intein, from a primary translation product. In SICLOPPS, split-intein\ndomains, comprising separately expressed N-terminal (IN) and C-terminal\n(IC) segments of an intein, reassemble within cells to form an active\nintein ( Scheme  \n ).\na When the two fragments\ncome\ninto proximity, they reconstitute the complete intein. The activity\nof the intein triggers acyl substitution reactions, leading to the\ncyclization of the peptide library and the release of the intein.\nThe first amino acid involved in the acyl substitution must be nucleophilic,\nsuch as cysteine or serine. If cysteine is present, a thioester intermediate\nforms, which subsequently reacts with the nucleophile at position\n1 of the extein (X = O or S). This reaction creates a lariat structure\nthat rearranges to yield a cyclic peptide.\nWhen the two fragments\ncome\ninto proximity, they reconstitute the complete intein. The activity\nof the intein triggers acyl substitution reactions, leading to the\ncyclization of the peptide library and the release of the intein.\nThe first amino acid involved in the acyl substitution must be nucleophilic,\nsuch as cysteine or serine. If cysteine is present, a thioester intermediate\nforms, which subsequently reacts with the nucleophile at position\n1 of the extein (X = O or S). This reaction creates a lariat structure\nthat rearranges to yield a cyclic peptide.\nThe process initiates with the creation of a library of target\npeptides, also known as exteins, flanked by the C-terminal and N-terminal\nsegments of a split intein (IC and IN, respectively), using conventional\nmolecular biology methods.  These fusion\nproteins undergo folding to activate the intein. To facilitate splicing,\nthe initial amino acid of the target peptide must be a nucleophilic\ncysteine or serine. However, there are no further restrictions on\nthe number or type of amino acids within the target peptide. This\nallows for the assembly of cyclic peptides of diverse sizes and sequences.\nIn SICLOPPS, the peptide of interest is initially synthesized as a\nlinear precursor with an N-terminal cysteine forming thioester. The\nthioester reacts with the nucleophile at position 1 of the extein\n(X = O or S), forming a lariat that rearranges to yield a cyclic peptide.\nTo create a plasmid library that encodes a diverse array of cyclic\npeptides, the extein sequence is modified using a degenerate oligonucleotide.\nThe number of variable amino acid positions in the library is influenced\nby the transformation efficiency of the host organism, typically  E. coli .  The degenerate oligonucleotide\nencodes the variable segment as repeats of NNS or NNB, where N signifies\nany of the four DNA bases (A, C, G, or T), S represents either C or\nG, and B indicates C, G, or T. The NNS and NNB sequences cover 32\nand 48 codons, respectively, including all 20 amino acids while excluding\nthe UAA and UGA stop codons from the library. The design and synthesis\nof the degenerate oligonucleotide carefully control the number of\nrandomized amino acids, as well as the inclusion of specific amino\nacids at designated positions, at the DNA level. Typically, 5 or 6\nvariable amino acids are introduced, ensuring that the total number\nof cyclic peptide library members (3.2 × 10 6  and 6.4\n× 10 7 , respectively) remains below the maximum number\nof  E. coli  transformants (typically 10 9 ), which guarantees that each member of the library can be assessed.\nWhile it is possible to generate and screen larger cyclic peptide\nrings with more randomized amino acid positions, the size of such\na library would still be limited by the transformation efficiency\nof the host organism. Historically, the trans-splicing split intein\nfrom DNA polymerase III (DnaE) derived from the cyanobacterium  Synechocystis sp . (Ssp) PCC6803 has been utilized in the\nSICLOPPS approach. However, inteins from  Nostoc punctiforme  (Npu) have shown faster splicing rates and better tolerance to amino\nacid substitutions near the splice junctions compared to Ssp inteins.\nDespite these advantages, some variants from the Npu SICLOPPS library\nwere found to be toxic to  E. coli . To mitigate this\nissue, a SsrA degradation tag was integrated into the Npu SICLOPPS\ninteins, enabling the bacterial protease ClpXP to degrade the spliced\ninteins.\nThis high-throughput screening platform has been used\nto discover\ncancer treatments, particularly for identifying cyclic peptides that\ninhibit the HIF-1α/HIF-1β protein–protein interaction.  HIF-1 is a heterodimeric transcription factor,\nand its role in angiogenesis, tumor growth, and metastasis is well\nestablished. In fact, the HIF-1α isoform is overexpressed in\nmany cancers, and its activation, along with oncogene activation and\nloss of tumor suppressor function, is associated with HIF-1 activation.\nA HIF-1 bacterial reverse two-hybrid system (RTHS) was developed and\nused to screen a plasmid-encoded SICLOPPS library of 6-mer cyclic\npeptides to inhibit the dimerization of HIF-1. From a library of 3.2\nmillion peptides, cyclo-CLLFVY was identified and proven to effectively\ninhibit the HIF-1α/HIF-1β protein–protein interaction\nboth in vitro and in cells.\nA more recent study combines SICLOPPS\nwith next-generation sequencing\n(NGS) and biopanning to identify novel cyclic hexapeptides targeting\ntumors.  The study presents a refined\nSICLOPPS screening method and workflow, incorporating pooled colony\ncollection, NGS, and biopanning, which improves screening accuracy\nand reduces false positives. Among the peptides identified, cyclo-CLLFCL\nexhibited the highest activity both in vitro and in cellular assays.\nAnother study demonstrated how the identified cyclic peptide interferes\nwith the Gag-TSG101 interaction, disrupting the complex and preventing\nHIV from budding out of the cell.  The\nGag-TSG101 interaction involves the binding of the HIV Gag protein\nto TSG101, a host cell protein. Because the peptide targets the host\nprotein, it is less likely to be circumvented by viral mutations,\nin contrast to treatments that target viral functions directly.\nWhile phage display\npeptide libraries offer extensive diversity, they may exhibit low\naffinity for the target and encounter issues related to the use of\nlive cells and phages. mRNA-encoded libraries have emerged as a promising\nalternative to overcome these limitations.\nmRNA display, a technique\nakin to phage display, is increasingly employed to discover new high-affinity\npeptide-based ligands for challenging therapeutic targets. It leverages\nsynthetic oligonucleotides and cell-free transcription/translation\nto generate large, naïve libraries of mRNA-barcoded peptides,\nenabling rounds of selection to identify high-affinity binders to\na protein target of interest. Initially developed in Nobel laureate\nJack Szostak’s lab, significant innovations have since been\nmade by Hiroaki Suga. \n ,\nThe RaPID (Random nonstandard\nPeptide Integrated Discovery) platform,\npioneered by Suga’s lab, represents a significant advancement\nin mRNA display technology. It introduces procedural improvements\nallowing for expedited selections within a week and incorporates unnatural\namino acids (UAAs) using robust RNA aptamers called flexizymes. RaPID\nintegrates mRNA display with a flexible in vitro translation (FIT)\nsystem, utilizing artificial flexible ribozymes to generate the desired\naminoacyl-tRNA.  This system allows for\nthe incorporation of any amino acid, natural or synthetic, expanding\nthe diversity of peptide sequences ( Figure  \n ).\n(a) RaPID\nleverages mRNA display technology and employs flexizymes\nto incorporate unnatural amino acids. This expands the diversity of\npeptides that cyclize spontaneously while still attached to their\nmRNA through puromycin (yellow sphere). (b) Cyclization occurs through\nthe formation of a thioester, generated by a spontaneous reaction\nbetween N-(chloroacetyl)-Tyr (depicted as a green sphere) and a cysteine\nresidue. Created in BioRender.\nRaPID enables the construction and screening of\nextensive libraries\nof cyclic peptides, offering a technologically advanced approach compared\nto conventional methods.\nIn a standard RaPID experiment, mRNAs\nconjugated with puromycin\nare expressed using the FIT system, encoding N-chloroacetylated (ClAc)\namino acids. Thioether macrocyclic peptides are generated by introducing\nunnatural N-(chloroacetyl)- d -Trp or N-(chloroacetyl)-Tyr\ninto mRNA-encoded libraries, followed by spontaneous cyclization with\nCys residues. The resulting products, cyclic peptides conjugated to\npuromycin, along with their respective mRNAs, are subjected to binding\naffinity assessment against target proteins using a systematic screening\napproach. \n , \n  Additionally, amino acid derivatives\nlike 5-hydroxytryptamine and benzylamine are synthetically assembled\ninto linear peptidic sequences, then cyclized using photogenic oxidative\ncoupling to yield fluorescent cyclic peptides.\nThe RaPID system enabled the identification of thioether-macrocyclic\npeptides with high affinity for the target protein. However, peptides\nproduced via this system are unprotected, which imposes constraints\non macrocyclization that must be chemo- and regioselective and occur\nunder mild, aqueous conditions. As a result, traditional macrolactonisation\nmethods commonly used in solid-phase peptide synthesis cannot be easily\napplied during ribosomal peptide synthesis. To overcome this limitation,\nthe Suga group developed an innovative approach for generating macrolactones\ndirectly within the context of ribosomal peptide synthesis.  This method involves incorporating the SPCG\nmotif into the peptide sequence. During the standard RaPID process,\ncysteine forms a self-acylating macrocycle, followed by serine forming\nan  O -acyl isopeptide through an intramolecular  S -to- O  acyl transfer. This post-translational\nmodification occurs spontaneously, producing cyclic depsipeptides\nin a one-pot reaction with variable sizes, ranging from 7 to 17 residues.\nThe study found that proline and glycine play a role in facilitating\nthe correct arrangement of residues for the acyl transfer. However,\nthe most critical factor is the positioning of serine and cysteine,\nas the SXCX motif is essential for the transfer process. For instance,\nthe CPSG motif was observed to be less efficient in facilitating this\nreaction.\nRaPID has transformed mRNA display into a powerful\ntool for identifying\npotent peptide inhibitors, leading to the establishment of successful\ncompanies like PeptiDream. While integrating unnatural amino acids\ninto biological libraries presents challenges, advancements in genetic\ntechnologies have facilitated the engineering of mRNAs and tRNAs.\nIn recent years, several biological cyclic peptide libraries incorporating\nunnatural amino acids have been reported. \n − \n \n  The thioether linkage, utilized in the RaPID system, holds significance\nin the development of cyclic mimetics containing Cys.\nMethods\nsuch as RaPID and other recombinant and enzymatic approaches are considered\ngreen technologies because they do not require the use of harmful\nsolvents or reagents, making them safer for both the environment and\nhuman health. However, in pharmaceutical companies, peptides are still\nsynthesized using chemical methods, often requiring a significant\nexcess of protected amino acid monomers, costly activation agents,\nharsh reagents, and large amounts of organic solvents. This approach,\nespecially at scale, generates considerable waste. \n , \n  While efforts have been made to develop greener synthetic processes,\nchallenges in achieving sustainable production and efficient purification\npersist.  The production of large peptides\nand proteins typically involves the synthesis of smaller peptide fragments,\nwhich are then coupled. In the pharmaceutical sector, the use of protected\npeptide fragments for coupling is a common strategy for therapeutic\npeptide production while condensation of unprotected fragments has\nproven less efficient and practical. \n ,\nRecent\nsuccesses in the development and market approval of long peptides\n(>30 residues) containing unnatural amino acids underscore the\nutility\nof chemical approaches, particularly hybrid processes, for minimizing\nimpurities, streamlining purification, and meeting stringent regulatory\nstandards. For example, the hybrid synthesis of Tirzepatide integrates\nsolid-phase peptide synthesis (SPPS) and liquid-phase peptide synthesis\n(LPPS), facilitating impurity control and purification.  Chemo-enzymatic peptide synthesis (CEPS) presents\nan alternative, leveraging enzymes for fragment condensation, as demonstrated\nby ligases. \n , \n  CEPS, a promising green alternative,\nemploys water-based conditions for fragment coupling instead of organic\nsolvents such as those used in LPPS.  Nonetheless, SPPS remains the primary method for fragment synthesis,\noften utilizing dimethylformamide as a solvent.\nBiocatalysis\nhas had a transformative impact on the synthesis of\nsmall molecules (e.g., sitagliptin), but its application to medium-sized\nmolecules like peptides and oligonucleotides has been comparatively\nlimited.  Enzymes such as sortases, butelases,\ntrypsiligases, and engineered variants of subtilisins like omniligases,\nsubtiligases, and peptiligases have been employed for peptide fragment\nligation, with significantly advancing CEPS. \n − \n \n  These enzymes facilitate the production of linear and cyclic peptides,\nprotein conjugates, and therapeutic peptides. \n − \n \n  Notably, omniligase-1, a broad-specificity ligase engineered from\nsubtilisin BPN’, was successfully used to synthesize exenatide,\nand peptiligase has been applied for gram-scale quantities of therapeutic\npeptides such as thymosin-α1, exenatide, and the kalata B1 variant\nT20K.  These results highlight the potential\nof CEPS for adoption in sustainable, large-scale manufacturing of\ntherapeutic peptides.\nTypically, CEPS employs Cam esters as\nacyl donors and catalyzes\nthe condensation of C-terminal peptide esters with N-terminal peptide\nfragments in aqueous conditions. This method minimizes hydrolysis\nby favoring the condensation reaction kinetically. Furthermore, the\nengineering of optimized enzymes significantly reduces the formation\nof side products, such as those arising from ester hydrolysis or unintended\ncoupling with other N-terminal amines present in the reaction mixture.\nPawlas et al. employed the CEPS method\nto synthesize exenatide,\nspecifically by preparing the fragments H-1–21-O-Cam-L-NH 2  and H-22–39-NH 2  through solid-phase peptide\nsynthesis and ligating them using omniligase-1.  Their study demonstrated that enzymatic ligation proceeds\nefficiently under physiological pH and in the presence of 10% acetonitrile\nas a cosolvent, working effectively with both crude and purified fragments.\nHowever, the carboxamidomethyl (O-Cam) linker exhibited limited stability\nat high temperatures. To address this, the aromatic 4-hydroxymethylbenzoic\nacid (HMBA) linker was evaluated as a more robust alternative. Using\nthis approach, the H-1–21-HMBA-K fragment was synthesized with\nhigh yield and purity and subsequently coupled to the H-22–39-NH 2  fragment on a large scale via omniligase-1 catalysis, yielding\n53 g of crude exenatide. The process was further assessed in terms\nof manufacturing cost, complete E factor (cEF), and carbon intensity\n(CI), comparing it to both conventional and lab-scale CEPS benchmark\nprocesses. The results revealed that the CEPS process employing the\nH-1–21-HMBA-K fragment was not only successfully scaled up\nbut also demonstrated significant improvements in economic efficiency\nand environmental sustainability compared to both benchmark methods.\nThe Cabri group reported another hybrid system based on green solid-phase\npeptide synthesis (GSPPS) for the preparation of peptide fragments,\ncombined with omniligase-1 for fragment coupling.  This approach was tested for the synthesis of liraglutide.\nInitially, two liraglutide fragments, H-(1–11)-CamFK-NH 2  and H-(12–31)–OH, were synthesized ( Figure  \n ).\nCEPS (Chemo-Enzymatic\nPeptide Synthesis) is an approach that combines\norganic chemistry strategies with enzymatic catalysis to form peptide\nbonds. Specifically, the peptide of interest (e.g., liraglutide, shown\nin the figure) is synthesized in fragments using solid-phase synthesis\nand green solvents. The crude fragments are then condensed enzymatically\nin an aqueous solution to produce the complete peptide. This method\noffers a more sustainable alternative to conventional SPPS in DMF\nand results in peptides with fewer impurities, thanks to the fragment\ncondensation process. Created in BioRender.\nThe first fragment featured a C-terminal activated\nwith the carboxamidomethyl\nester (OCam) and was extended by two amino acids, phenylalanine and\nlysine. The addition of the two amino acids following the Cam ester\nmoiety improved both the solubility and substrate interaction with\nthe enzyme, further enhancing coupling efficiency. However, the use\nof the OCam ester significantly reduced atom economy as the -OCam-FK-NH 2  fragment is not present in the final product. Additionally,\nintroducing -OCam-FK requires more solvent and reagent, negatively\nimpacting the overall greenness of the process. This limitation further\nemphasized the need to replace DMF with greener alternatives. The\ngroup evaluated various green solvents and their combinations, identifying\nN-butylpyrrolidone/dimethyl carbonate (8:2) as the most effective\nmixture. This combination provided good results compared to standard\nconditions with DMF, yielding higher purity and greener process metrics.\nTo further enhance process efficiency, fragment ligation was performed\non crude peptides. This strategy, inspired by previous work on CEPS,\ndemonstrated significantly lower process mass intensity (PMI) metrics\ncompared to approaches using purified fragments. The ligation reaction\nwas monitored by HPLC, and after 24 h, it resulted in an 81% yield\nin solution.\nBiocatalysis holds significant potential for peptide\ncyclization.\nHowever, commercially available cyclases are limited in their ability\nto cyclize peptides smaller than 10 amino acids. This limitation underscores\nthe need to investigate alternative nonribosomal cyclases, particularly\nthose capable of cyclizing natural peptides with scaffolds ranging\nfrom 4 to 15 residues. Among these, the SurE cyclase, a key enzyme\nin the surugamide biosynthetic pathway from various Streptomyces species,\nis of particular interest due to its remarkable substrate tolerance,\nmaking it a promising candidate for biocatalytic applications.\nThe SurE enzyme can be employed in combination\nwith the CuAAC reaction\nto produce bicyclic peptides. After the enzyme catalyzes the head-to-tail\ncyclization, the subsequent click reaction between the azide and alkyne,\nintroduced immediately after the enzymatic step, can occur. It was\nsufficient to add a copper-based catalyst and ascorbic acid directly\ninto the enzymatic reaction mixture. This two-step cyclization proceeds\nin a one-pot reaction without the need to purify the monocyclic intermediate.\nThis chemoenzymatic strategy facilitated the efficient synthesis of\nbicyclic peptides containing hexa-, octa-, and undecapeptidyl head-to-tail\ncyclic scaffolds.\n\nA category of peptidomimetics comprises molecules devoid of a peptide\nmain chain, featuring only side chains. These structures are commonly\nemployed to emulate secondary structures. However, their lack of a\npeptide backbone renders them highly flexible. As a result, there\nis no distinct global minimum energy state corresponding to a specific\nsecondary structure. Instead, they resemble various secondary conformations\nconcurrently and can adapt to diverse binding scenarios. For instance,\nthey may occupy compact enzyme cavities typically inaccessible to\nother peptides or peptidomimetics with a backbone, especially when\nthe precise binding conformation is unknown.  The key consideration in designing such molecules is to avoid high\nthermodynamic costs and insurmountable kinetic barriers, ensuring\neasy obtainment of the desired structures. Therefore, their backbone\ndesigns must incorporate moieties restricting degrees of freedom.\nThese peptidomimetics are alternatively\ntermed minimalistic or\nuniversal mimics, reflecting the absence of a backbone or the capability\nto adopt any secondary structure. Pioneering examples were introduced\nby Hirschmann and Smith, who designed β-turn analogues. Their\napproach involved incorporating additional molecules such as sugars,\ncatechols, and steroids to position significant side chains at appropriate\ndistances, elucidating their activity. \n , \n  The Hamilton\ngroup proposed minimalist helical mimics, utilizing terphenyl scaffolds\nto present side chains in optimal orientations. In contrast to the\nstructures suggested by Hirschmann and Smith, these helical mimetics\nexhibit sufficient rigidity.\nAdditional\nexamples of minimalist peptidomimetics include pyrrolinone-pyrrolidine\noligomers derived from tetramic acids as critical starting materials.\nThese mimetics, featuring two noncontiguous side chains, can adopt\nthe conformation of three different helix types and both parallel\nand antiparallel β-sheets. The mimetic with three noncontiguous\nside chains predominantly assumes antiparallel β-sheet conformations.\nWhen paired with Val, cyclophanes serve\nas minimalist cyclic peptidomimetics.\nVal-cyclophanes exhibit self-assembly into an organized architecture\nbased on a fibrillar network. The resulting supramolecular network\nis assembled through physical interactions and entraps a diverse range\nand substantial amounts of solvents, forming robust gels.\n\nObesity and overweight pose a significant global health concern,\naffecting millions of people, including both adults and children.\nObesity is strongly linked with dyslipidemia, characterized by elevated\nblood levels of low-density lipoprotein (LDL) and cholesterol, and\nis also a major risk factor for type 2 diabetes mellitus (T2DM).  Historically, type 2 diabetes has been treated\nwith metformin, which remains a first-line therapy alongside sulfonylureas,\nthiazolidinediones, and insulin. \n − \n \n  Insulin is also a medicine\ncommonly used to control glucose levels. Insulin is a peptide hormone\ninitially extracted from animal sources, was later produced through\nrecombinant DNA technology, yielding a safer product that minimized\npatient allergic reactions. Advances in genetic engineering enabled\nspecific amino acid modifications to enhance the ADMET properties\nof insulin.  Notably, researchers have\nengineered insulins that avoid hexamer formationan inactive\nstorage form in the bodyand developed both fast-acting (lispro,\naspart, glusine) and slow-release (glargine, detemir, deglutec) insulin\nanalogues, offering various therapeutic options for patients. Insulin\nis also effective in managing type 1 diabetes ( Figure  \n ).\nStructure\nof insulin, consisting of two chains, is shown at the\ntop. Several engineered insulin analogues have been developed through\namino acid modifications or additions. In the figure, green spheres\nindicate residues in the original insulin sequence that have been\nmodified. Most engineered insulins involve modifications to residues\non the B-chain, particularly at positions 3, 28, 29, and 30. For example,\nposition 29 has been modified with the addition of a lipid chain,\nas in detemir and degludec, which exhibit long-lasting action. Degludec\nalso lacks threonine at position 30. Long-lasting action can also\nbe achieved by adding arginine residues to the B-chain, extending\nit to 32 residues, as in glargine. By modifying the charges at positions\n3, 28, and 29 of the B-chain, fast-acting analogs such as lispro,\naspart, and glulisine can be developed. Created in BioRender.\nAmlyn, a peptide hormone cosecreted with insulin\nby pancreatic\nβ-cells in response to meals, is present at low levels in type\n1 diabetes patients but elevated in those with type 2 diabetes. An\namylin analogue, pramlintide, was approved in 2005; \n , \n  it is coadministered with insulin at mealtime, often in combination\nwith metformin and/or sulfonylureas. Human amylin is highly amyloidogenic,\nbut studies showed that rat amylin, which includes proline residues,\ndoes not readily form amyloid aggregates. As a result, prolines were\nsubstituted for Ala25, Ser28, and Ser29 in human amylin to develop\nthe pramlintide analog.\nHowever,\nmaintaining glucose homeostasis can be challenging for\ncertain patients, necessitating new therapeutic targets and more effective\ntreatments. A major advancement in diabetes treatment was achieved\nwith the development of glucagon-like peptide-1 receptor agonists\n(GLP-1RAs). GLP-1RAs mimic the action of the natural peptide GLP-1\nby binding to the GLP-1 receptor (GLP-1R), thus producing effects\nsimilar to those of the endogenous peptide. GLP-1 is a peptide hormone\nbelonging to the incretin family, released in two phases: an initial\nphase approximately 10–15 min postmeal, followed by a secondary\nrelease 30–60 min later from intestinal L-cells. Its active\nforms include GLP-1(7–37) and amidated GLP-1(7–36).  GLP-1 receptors are present on the membranes\nof various cell types, enabling GLP-1 and its agonists to exert multiple\neffects throughout the body, impacting not only the digestive system\nbut also the brain, heart, kidneys, and muscles ( Figure  \n ).\nGLP-1 exerts multiple effects by binding to its receptor, which\nis expressed on the membranes of various cell types. GLP-1 receptors\nare found in the pancreas, muscle tissue, gastrointestinal tract,\nbrain, heart, kidneys, and adipose tissue. GLP-1 helps regulate blood\nglucose levels by stimulating insulin release, promoting glucose storage\nas glycogen, and increasing satiety. Additionally, it has cardiovascular\nbenefits, such as lowering blood pressure, and plays a role in combating\nobesity. Created in BioRender.\nIn the gastrointestinal tract, GLP-1 enhances insulin\nsecretion\nfrom pancreatic β-cells, triggers somatostatin release from\nδ-cells, inhibits glucagon release from α-cells, and slows\ngastric emptying. This leads to improved blood glucose regulation\nand promotes satiety, which assists in weight loss ( Figure  \n ).\nWhen nutrients enter\nthe small intestine, L-cells synthesize and\nsecrete GLP-1, which is proteolytically processed to yield the active\nforms GLP-1(7–37) and GLP-1(7–36)­NH 2 . Active\nGLP-1 binds to its receptor on the cell membranes of pancreatic β-cells.\nThis receptor is a G protein-coupled receptor, and its interaction\nwith GLP-1 triggers an intracellular signaling cascade that leads\nto the synthesis of insulin. Insulin is then secreted via vesicles\nthat fuse with the plasma membrane. Created in BioRender.\nThis glucose-dependent insulinotropic effect means\nthat GLP-1 actions\nare triggered only when blood glucose levels are elevated above normal\nfasting plasma levels. This is particularly beneficial for diabetes\nmanagement as it reduces the risk of hypoglycemiaa common\nside effect of several antidiabetes drugs, including insulin.  Additionally, GLP-1 demonstrates positive effects\non multiple tissues and organs, with the widespread presence of GLP-1R\nsuggesting that GLP-1 plays broader roles beyond glucose metabolism.\nSignificant efforts were made to develop GLP-1 as a therapeutic drug\nafter researchers observed that intravenous injections of GLP-1 had\nbeneficial effects on insulin secretion and blood glucose control\nin patients with type 2 diabetes.\nStructure–activity studies using alanine-scanning have revealed\nthat residues His7, Gly10, Phe12, Thr13, Asp15, Phe28, and Ile29 are\ncritical for GLP-1 receptor interaction, with the active form identified\nas GLP-1(7–37).  The natural active\nGLP-1(7–37), hereafter referred to simply as GLP-1 in this\narticle, is rapidly degraded by dipeptidyl peptidase 4 (DPP-4), which\ncleaves between residues Ala8 and Glu9, and is swiftly cleared renally\nwithin 1–2 min.  Substituting\nthe position-8 residue improves DPP-4 resistance, though the peptide\nremains susceptible to rapid renal clearance. DPP-4 is the primary\nenzyme responsible for GLP-1 inactivation. Replacing Ala8 with the\nα,α-dimethyl amino acid Aib (α-aminoisobutyric acid)\nor Gly has effectively prevented unwanted proteolytic cleavage.\nIn the early 1990s, a GLP-1 analogue was discovered in the venom\nof the Gila monster. This peptide, exendin-4, exhibited 53% sequence\nhomology with human GLP-1 and proved highly stable against DPP-4 degradation\nand resistant to renal clearance in humans.  Exendin-4 contains glycine at position 8, replacing alanine and\navoiding DPP-4 cleavage. It also has a tail of proline, alanine and\nserine residues (PASylation;  Section  \n ) followed by three proline residues that\nform a steric shield around the peptide, reducing protease accessibility.\nSpecifically, this tail forms a “Trp cage” at the C-terminus,\nwhich protects the peptide from degradation by another protease, neutral\nendopeptidase (NEP). \n , \n  The C-terminal tail also increases\nthe peptide’s molecular size and hydrodynamic radius, reducing\nrenal filtration and extending its half-life to approximately 2.4\nh.\nExendin-4 acts as a GLP-1 receptor agonist and served as\nthe foundation\nfor the pharmaceutical formulation marketed as exenatide, sold under\nthe brand name Byetta by AstraZeneca in 2005 and later by Bristol-Myers\nSquibb. The sequence of exenatide is identical to that of exendin-4.\nThe drug is administered to adults twice daily via subcutaneous injection\nbefore main meals.  In 2017, a long-acting\nformulation, Bydureon BCise, was approved, enabling once-weekly administration.  Further advancements have been made in developing\nGLP-1RAs. Currently, the FDA has approved seven different GLP-1RAs\nfor the treatment of type 2 diabetes and obesity ( Figure  \n ,  Table  \n ).\nNative GLP-1\nsequence contains essential amino acids (green) that\nare critical for its interaction with the receptor. In designing GLP-1\nanalogues, efforts have been made to preserve these essential residues\nwhile modifying less critical ones. Modifications relative to the\nnative sequence are shown in orange. A key modification common to\nall analogues, except liraglutide, is at position 8, where alaninea\ntarget for proteolytic cleavagehas been replaced with Aib\nor Gly to enhance stability. Additional changes have been made to\ninternal residues or at the termini, such as in exenatide and lixisenatide,\nwhich feature added residues. Residues left unchanged are shown in\nwhite, while those in blue represent spacers attached to the side\nchain of lysine 26 to link lipid moieties, as seen in liraglutide\nand semaglutide. In contrast, albiglutide and dulaglutide incorporate\nproteins at the C-terminus: albumin in the case of albiglutide and\nan antibody for dulaglutide. Both albiglutide and dulaglutide feature\ntwo copies of the GLP-1 sequence. Tirzepatide is unique, as it combines\nkey sequences from GLP-1 and GIP, enabling the analogue to bind to\nboth receptors and provide more effective treatment for diabetes and\nobesity. Tirzepatide also includes a lipid tail to increase the circulation\ntime of the drug. Created in BioRender.\nOne of these GLP-1RAs is liraglutide, approved\nin 2010 and marketed\nby Novo Nordisk in two formulations: Victoza for diabetes treatment\nand Saxenda for weight loss.  In 2014,\ndulaglutide, marketed by Lilly as Trulicity, received FDA approval.  Also approved in 2014, albiglutide was developed\nby GSK and marketed as Tanzeum; however, it was later withdrawn from\nthe market for commercial reasons.  Lixisenatide,\nproduced by Sanofi, was approved in 2013 and sold as Lyxumia in Europe.  In 2016, it was also approved under the brand\nname Adlyxin in the United States, though it was later discontinued\nfor business reasons.  In 2017, semaglutide\nreceived FDA approval and was marketed by Novo Nordisk as Ozempic,\nwith an additional oral formulation approved in 2019 as Rybelsus and\na version for weight management called Wegovy approved in 2021. \n − \n \n  Semaglutide has shown superior efficacy compared to liraglutide.\nIn 2022, Lilly received FDA approval for tirzepatide, marketed as\nMounjaro.\nIt is established that\nincreasing the molecular weight of peptides\nis crucial in drug development, as steric hindrance enhances stability\nagainst degradation. Additionally, the larger size reduces renal clearance,\nthereby prolonging plasma circulation time. In the development of\nnew GLP-1 receptor agonists, researchers have recognized that increasing\nthe molecular weight of peptides can be achieved through various strategies\nand molecules. Lixisenatide is a modified form of exenatide, featuring\na longer peptide chain composed of 44 amino acids, with the C-terminal\nproline residues replaced by six lysines. This modification has demonstrated\nan increased affinity for the GLP-1 receptor compared to both exenatide\nand native GLP-1, and lixisenatide can be administered once daily.  The lysine tail enhances receptor binding affinity;\nhowever, it does not significantly extend the half-life of lixisenatide,\nwhich is approximately 3 h, comparable to exenatide’s half-life\nof 2.4 h. Although amino acid addition and substitution effectively\nreduce proteolytic degradation, modified GLP-1 analogues continue\nto face rapid renal clearance, which limits efforts to extend peptide\ncirculation time. Consequently, strategies beyond adding amino acids,\nsuch as those employed in exenatide, have been explored. These include\nlipidation and conjugation to larger proteins, which may enhance stability\nand prolong the circulation time of the peptides in the bloodstream\n( Section  \n ).  Lipid conjugation has been employed in the\ndevelopment of long-acting analogues such as insulin detemir, insulin\ndegludec, liraglutide, semaglutide, and tirzepatide, allowing for\nadministration either once daily (as seen with detemir, degludec,\nand liraglutide) or once weekly (as with semaglutide and tirzepatide).\nThe incorporation of lipids not only increases the peptide’s\nsize but also enhances its binding to albumin, significantly improving\nplasma circulation compared to the addition of the proline, serine,\nand alanine tail in exenatide. Albumin acts as a protective shield\nfor the peptide, preventing protease degradation. Furthermore, lipid\nconjugation provides additional benefits, including delayed release\nfrom the injection site and reduced immunogenic response.  In the case of liraglutide, it has been observed\nthat the lipid tail promotes the aggregation of the peptide into hexa-,\nhepta-, or octamers, remaining in this oligomeric form under specific\npH or ionic strength conditions.  When\nthese conditions change, the oligomers dissociate into monomers, which\nenter circulation and bind to albumin noncovalently. This aggregation\nmechanism is responsible for the slow release of the peptide at the\ninjection site. Overall, incorporating the lipid significantly enhances\nthe pharmacokinetics of the peptide.\nLike all peptides, GLP-1\nanalogues have a preferred orientation\nfor binding to their receptor. The N-terminus contains critical residues\nnecessary for receptor activation, and attaching the lipid tail to\nthis end significantly reduces activity.  Consequently, hydrophobic components are typically conjugated to\nthe C-terminus, with a spacer used between the lipid and the peptide\nto enhance flexibility. For liraglutide, the spacer is γ-glutamic\nacid at position Lys26, while the lipid component is palmitic acid.\nIn semaglutide, the lipid tail is longer, consisting of 18 carbon\natoms and featuring two carboxylic acid groups. The spacer in semaglutide\nincludes one γ-glutamic acid and two 8-amino-3,6-dioxaoctanoic\nacid (ADO) units.  Both liraglutide and\nsemaglutide maintain the same sequence as natural GLP-1. Their distinguishing\ncharacteristics compared to the native peptide include modifications\nat positions 26 and 34, where lysine is conjugated to a lipid chain\nor substituted by arginine, and at position 8 in semaglutide, where\nan Aib is incorporated to enhance stability against DPP-4 degradation.\nAnother successful strategy involves\nthe conjugation of peptides\nto proteins, as exemplified by albiglutide and dulaglutide. Albiglutide\nis linked to human albumin, while dulaglutide is covalently attached\nto a fragment of human IgG4. \n , \n  These analogues consist\nof two peptide chains with sequences similar to natural GLP-1, incorporating\na few mutations to enhance stability, particularly at position 8,\nwhere glycine replaces alanine.\nIn dulaglutide, the two peptide\nchains are identical and truncated,\nlinked covalently to the antibody fragment via a flexible linker.\nTheir stability is ensured by a disulfide bridge formed between the\ntwo IgG4 Fc regions. In contrast, albiglutide also contains two peptide\ncopies; however, they are arranged sequentially, with conjugation\nto human albumin occurring solely at the C-terminus.\nAnother approach to improve circulation half-life\nof GLP-1 agonists\ninvolves the use of carriers such as polymeric hydrogels, nanoparticles,\nor microparticles. However, when it comes to peptides, only poly­(lactic- co -glycolic acid) (PLGA) has been approved by the FDA, which\nconsiders it a safe and effective carrier for therapeutic peptides.\nPLGA is biocompatible and is gradually degraded by the body, allowing\nfor prolonged drug release.  PLGA has\nalso been employed in formulations for antidiabetic peptides. For\ninstance, in a collaboration between Lilly, Amylin, and Alkermes,\nPLGA microspheres were developed to deliver exenatide. Although a\nPLGA formulation capable of releasing the drug over several months\nis theoretically possible, the exenatide formulation, marketed as\nExenatide QW (Bydureon) by AstraZeneca, was limited to a once-weekly\ninjection.  Upon subcutaneous administration,\nexenatide is initially released from the surface and surface pores\nof the microspheres during the first 48 h. The product was designed\nfor a slow initial release phase to minimize adverse effects such\nas nausea and vomiting.  The second phase\ninvolves the gradual diffusion of the drug from the polymer matrix,\nwith peak plasma concentrations observed after approximately 2 weeks.  In the third phase, degradation of the microparticles\noccurs. Overall, drug release with this technology lasts for about\n11 weeks. Before administration, PLGA microparticles must be suspended\nin a phosphate buffer. Subsequently, the Bydureon Bcise device was\ndeveloped, in which PLGA particles are suspended in triglycerides\nand delivered via a single-dose autoinjector.\nAs noted, one limitation of PLGA microparticles in\nthe Bydureon\nformulation is the inability to achieve a consistent release rate\nof GLP-1RAs. To address this issue and provide steady, continuous\ndelivery of GLP-1RAs, other controlled-release devices such as osmotic\npumps have been explored. A representative product is ITCA 650, developed\nby Intarcia Therapeutics.  ITCA 650 is\nan implantable subdermal osmotic titanium mini-pump designed for the\ncontinuous release of exenatide over a period of up to six months.\nIn 2024, the FDA rejected the New Drug Application (NDA) for ITCA\n650 following a unanimous vote by an FDA advisory committee, which\nraised significant safety concerns. These concerns included potential\nrisks of acute kidney injury and cardiovascular side effects. Additionally,\nthe delivery system, which operates on a continuous release mechanism,\nwas criticized for its inconsistent drug release, further jeopardizing\npatient safety. This rejection marks the third setback Intarcia company\nhas encountered regarding ITCA 650.\nThe discussion surrounding tirzepatide is distinct, as it functions\nas a dual agonist by combining the actions of two incretins: gastric\ninhibitory polypeptide (GIP) and GLP-1 ( Figure  \n ).\nTirzepatide features a sequence homologous\nto both GIP and GLP-1,\ngiving the peptide dual action by targeting two receptors. The amino\nacids at the C-terminus enhance stability, while the lipid tail prolongs\ncirculation time, improving its pharmacokinetic profile. Created in\nBioRender.\nGIP, like GLP-1, stimulates insulin secretion.\nThe dual agonist\neffect of tirzepatide appears to confer a superiority over semaglutide,\nevidenced by a greater reduction of more than 2% in glycated hemoglobin\n(HbA1c), a marker of chronic hyperglycemia.  Tirzepatide comprises 39 amino acids and is based on the GIP sequence\nwith several modifications. Similar to semaglutide and liraglutide,\nit contains a lipid tail, specifically a C-20 fatty acid (1,20-eicosanedioic\nacid) linked to lysine at position 26 via a spacer moiety. Additionally,\nit incorporates two Aib residues at positions 8 and 19, along with\namidation at the C-terminus.\nVarious synthetic strategies for\nthe production of GLP-1 have been\nreported. These approaches can be broadly classified into two main\ncategories: recombinant techniques and fully synthetic strategies.\nA recombinant strategy employed by Novo Nordisk involves the production\nof liraglutide through recombinant DNA techniques, followed by the\nin vitro attachment of a γ-(Pal-Glu-O t Bu) moiety\nto Lys26. In this approach, the peptide is initially synthesized as\na precursor with an N-terminal extension, which serves multiple functions:\nit protects the precursor molecule from proteolytic degradation within\nthe host cell or culture medium, facilitates purification, and minimizes\nfibril formation. Following expression and purification, the N-terminal\nextension is removed, exposing Lys26. Under controlled in vitro conditions,\na lipid chain, specifically a palmitic acid moiety, is subsequently\nconjugated to Lys26 via a glutamic acid linker.  To ensure that lipid chain attachment occurs exclusively\nat Lys26 and does not affect Lys34, the latter was replaced with arginine.\nThis substitution prevents nonselective lipidation, ensuring site-specific\nmodification of the peptide. Another approach to obtain liraglutide\ninvolves the use of fusion peptides. A fusion peptide consists of\nthree components: the target peptide, an affinity tag, and a cleavable\ntag. Following recombinant production, the peptide is purified using\naffinity chromatography. In the final step, the cleavable tag is removed,\nyielding the purified peptide.\nRecombinant techniques have also been employed for the production\nof albiglutide and dulaglutide. These two GLP-1 analogs are large\nmolecules composed exclusively of naturally occurring amino acids,\nmaking in vivo expression the most suitable approach for their production. \n , \n  For the synthesis of the dulaglutide dimer, the Fc portion of IgG4\nwas modified to introduce serine residues. Following protein synthesis,\nthe two Fc chains spontaneously dimerize through the formation of\ndisulfide bonds, resulting in the final dimeric structure.\nIn\nchemical synthesis, challenges such as peptide aggregation and\nthe presence of numerous deletion peptides, which often coelute with\nthe target peptide, are commonly encountered.  The branched structure, combined with fatty acid modifications\nand a distinct amino acid sequence, promotes peptide folding and aggregation,\nmaking the chemical synthesis of high-purity liraglutide and semaglutide\nparticularly challenging. This complexity was one of the key reasons\nwhy recombinant approaches were initially favored for industrial production.\nIn general, research on chemical synthesis has focused on optimizing\nreaction conditions through the use of efficient and selective coupling\nreagents, orthogonal protecting groups, specialized resin linkers,\nand the incorporation of pseudoprolines and depsipeptide intermediates\ninto the peptide sequence.\nSome orthogonal protecting groups\nemployed in the total chemical\nsynthesis of liraglutide and semaglutide are Alloc, Mtt, or ivDe on\nthe Lys residue to enable selective modification. \n − \n \n \n \n  With this approach, at the end of the SPPS process, the protecting\ngroup on the lysine residue is removed, followed by the coupling of\nN α -protected Glu-O t Bu. Subsequently,\nN α -deprotection is performed, allowing for the final\nconjugation of Pal–OH to complete the synthesis ( Scheme  \n ).\na Following SPPS,\nthe protecting\ngroup on the lysine side chain is selectively removed, allowing for\nthe subsequent coupling of additional amino acids and the attachment\nof a lipid tail to complete the molecule.\nFollowing SPPS,\nthe protecting\ngroup on the lysine side chain is selectively removed, allowing for\nthe subsequent coupling of additional amino acids and the attachment\nof a lipid tail to complete the molecule.\nAnother approach for liraglutide involves the solution-phase synthesis\nof the dipeptide Fmoc-Lys­(Pal-γ-Glu-OtBu), which contains lysine\nand glutamate modified with a palmitoyl chain. This dipeptide is then\nincorporated during SPPS. Compared to the previous method, this strategy\nrequires fewer protecting groups. Specifically, Pal–OH reacts\nwith H-Glu-O t Bu to form Pal-Glu-O t Bu, which subsequently reacts with Fmoc-Lys-OH, yielding the dipeptide\nwith high purity ( Scheme  \n ). The only protective group used in this method is O t Bu on the carboxyl group of glutamate.\nThe use of copper­(II) lysinate offers a significant\nsimplification\nin the preparation of palmitoylated intermediates. Copper­(II) complexes\nof trifunctional amino acids, such as Lys, Asp, and Glu, can serve\nas temporary protecting groups, enabling the selective introduction\nof modifications into the side chain ( Scheme  \n ). The potential application of copper­(II)\nlysinate has been explored for the synthesis of lipidated intermediate\nbuilding blocks, which can subsequently be incorporated into the amino\nacid sequences of liraglutide and semaglutide. This strategy eliminates\nthe need for orthogonally protected lysine as a starting material,\nmaking it particularly advantageous for the industrial-scale production\nof peptides.\na Copper­(II) coordinates\nwith\nlysine, serving as a temporary protecting group. This strategy eliminates\nthe need for pre-synthesized orthogonally protected lysine derivatives,\noffering a more efficient and flexible approach to site-specific modification.\nCopper­(II) coordinates\nwith\nlysine, serving as a temporary protecting group. This strategy eliminates\nthe need for pre-synthesized orthogonally protected lysine derivatives,\noffering a more efficient and flexible approach to site-specific modification.\nOther approaches have integrated solid-phase\npeptide synthesis\n(SPPS) and liquid-phase peptide synthesis (LPPS) for the production\nof liraglutide. These methods involve the stepwise synthesis of peptide\nsegments via SPPS, followed by their coupling in solution. The SPPS/LPPS\nhybrid approach represents a promising alternative for minimizing\nthe formation of impurities, such as truncated sequences or peptides\nmissing one or more amino acids. The condensation site is typically\nselected based on the presence of amino acids that do not undergo\nepimerization during coupling. These include residues such as glycine\nor proline at the C-terminal position of the fragments, which help\nmaintain the stereochemical integrity of the peptide during fragment\nassembly. In a study conducted by the Cabri group, liraglutide was\nsynthesized using three peptide fragments: residues 7–16, 17–24,\nand 27–36.  The combined solid-phase/solution-phase\nstrategy followed the assembly order: 7–16 + [17–24\n+ (25–36 + 37)]. However, the 7–16 segment exhibited\nhigh hydrophobicity, resulting in significant solubility issues and\naggregation, making its use less efficient. A more effective strategy\nwas found to be 7–22 + (23–36 + 37), which improved\nsolubility and facilitated peptide assembly.\nA subsequent approach\ninvolves the incorporation of a pseudoproline\nresidue at the site of fragment condensation. Pseudoprolines are cyclic\nderivatives of Ser, Thr or Cys that form oxazolidine or thiazolidine\nrings, preventing the formation of the oxazolone intermediate, thereby\nreducing the risk of epimerization during peptide bond formation.  Additionally, pseudoproline residues play a\ncrucial role by effectively suppressing peptide aggregation.\nIn the case of liraglutide, as demonstrated by Cabri et al., a\npseudoproline residue was introduced between threonine-13 and aspartic\nacid-15, in the place of serine-14 ( Figure  \n ).  This modification\nled to an almost complete conversion to the target peptide, achieving\nexcellent purity and a high yield. However, pseudoproline residues\nlinked to a resin via a trityl-type linker exhibit a high propensity\nfor intramolecular cyclization, leading to diketopiperazine formation\nafter Fmoc removal using 20% piperidine in DMF ( Figure  \n ). The use of a piperidine/DBU/DMF\nmixture was explored to mitigate dipeptide detachment from the solid\nsupport. Despite this adjustment, the strategy was ineffective, resulting\nin a low yield of only 23%. To improve fragment solubility, 1% TritonX\nwas added to the condensation reaction, which was successfully completed\nwithin 3.5 h. Conversely, increasing the reaction temperature to 60\n°C did not yield favorable results, likely due to ester inactivation.\n(a)\nThe structure of the pseudoproline residue introduced between\nthreonine-13 and aspartic acid-15, replacing serine-14. (b) The diketopiperazine\n(DKP) formation of the pseudoproline dipeptides on the chlorotrityl\nchloride (CTC) resin is depicted. During SPPS, pseudoproline residues\ncan promote intramolecular cyclization, leading to DKP byproducts\nwhen the N-terminal amine of a growing peptide chain reacts with the\ncarbonyl group of the adjacent pseudoproline residue, particularly\nif the peptide-resin attachment is labile.\nThe SPPS/LPPS hybrid process was initially used\nfor the synthesis\nof enfuvirtide and later applied to tirzepatide. \n , \n  For the production of tirzepatide, four fragments were selected,\nand the disconnection points were chosen based on the potential for\nthe epimerization of the amino acid at the C-terminal of each fragment.\nEach fragment was obtained with a purity of approximately 98.5% and\ncondensate using PyOxim/ i Pr 2 Net or HATU/ i Pr 2 Net in DMSO/ACN ( Scheme  \n ).\na Disconnection points\nwere\ncarefully chosen based on the risk of epimerization at the C-terminal\namino acid of each fragment. Each fragment was synthesized with a\npurity of approximately 98.5% and coupled using PyOxim/iPr 2 NEt or HATU/iPr 2 NEt (also known as DIEA). \n b Each fragment was used at 1 equiv,\nwith iPr 2 NEt at 4.0 equiv and PyOxim at 1.4 equiv. The\ncoupling reaction was carried out for 3 h at 20 °C. Following\nthe coupling, the Fmoc protecting group was removed using 20 equiv\nof diethylamine (HNEt 2 ) for 1 h. The final fragment was\ncoupled using HATU (2.0 equiv) instead of PyOxim. Created in BioRender.\nDisconnection points\nwere\ncarefully chosen based on the risk of epimerization at the C-terminal\namino acid of each fragment. Each fragment was synthesized with a\npurity of approximately 98.5% and coupled using PyOxim/iPr 2 NEt or HATU/iPr 2 NEt (also known as DIEA).\nEach fragment was used at 1 equiv,\nwith iPr 2 NEt at 4.0 equiv and PyOxim at 1.4 equiv. The\ncoupling reaction was carried out for 3 h at 20 °C. Following\nthe coupling, the Fmoc protecting group was removed using 20 equiv\nof diethylamine (HNEt 2 ) for 1 h. The final fragment was\ncoupled using HATU (2.0 equiv) instead of PyOxim. Created in BioRender.\nLilly has also developed retatrutide, a triple\nagonist which is\ncurrently undergoing phase 2 clinical trials (ClinicalTrials.gov identifier:\n NCT04881760 ). Retatrutide is a single protein linked to a fatty diacid\nmoiety that activates the GIP, GLP-1, and GCG receptors, with the\nGCG receptor being associated with glucagon signaling.  In cell culture studies, retatrutide displayed\nlower potency than the natural ligands of the GCG and GLP-1 receptors\n(0.3 and 0.4 times as active, respectively), but exhibited significantly\nhigher potency at the GIP receptor (8.9-fold increase). The pharmacokinetics\nare dose-proportional, with an estimated half-life of around 6 days,\nmaking it suitable for once-weekly subcutaneous injection. In a phase\n2 study involving obese individuals without type 2 diabetes, retatrutide\nled to a weight reduction of up to 24.2% after 48 weeks.  The treatment also improved blood pressure,\nlipid profiles, and glycemic control. In an additional study, the\naim was to assess the mean relative change in liver fat (LF) from\nbaseline at 24 weeks among participants with metabolic dysfunction-associated\nsteatotic liver disease and at least 10% liver fat content. The observed\nmean changes in LF at 24 weeks were: – 42.9% (1 mg), –\n57.0% (4 mg), – 81.4% (8 mg), – 82.4% (12 mg), and +0.3%\nfor the placebo group. The reductions in liver fat were strongly associated\nwith decreased body weight, abdominal fat, and improvements in metabolic\nindicators tied to better insulin sensitivity and lipid metabolism.\nThere are significant concerns about the long-term use of GLP-1RAs,\nparticularly regarding the risk of pancreatitis and the potential\ndevelopment of pancreatic cancer. Animal studies have shown that exendin-4\ncan cause expansion of pancreatic duct glands and exacerbate chronic\npancreatitis.  Clinical data indicate\nthat patients treated with exenatide had a 6-fold higher incidence\nof pancreatitis compared to those using other antidiabetic medications,\nsuch as rosiglitazone, nateglinide, repaglinide, and glipizide.  However, other studies have found no evidence\nof a link between exenatide use and pancreatic injury in animal models.  Due to these inconsistent findings, there is\ncurrently no definitive conclusion regarding the association between\nexenatide use and serious adverse pancreatic events.\nSimilarly,\nliraglutide has been associated with an increased risk\nof thyroid cancer in rodent studies. Long-term activation of GLP-1\nreceptors was observed to stimulate calcitonin secretion and induce\nC-cell hyperplasia, leading to a higher incidence of medullary thyroid\ncancer in mice.  However, in human studies,\nliraglutide did not significantly affect calcitonin secretion.  As a result, the potential correlation between\nGLP-1RAs and thyroid cancer remains unclear.\nA frequently reported\nside effect of semaglutide treatment is nausea.\nAn alternative nonpeptidic drug has been introduced for weight management:\nMK-801 (also known as dizocilpine), an NMDA receptor antagonist. Prolonged\nsystemic administration of MK-801 induces anorexia and weight loss\nin rodents, but it is also associated with severe adverse effects,\nsuch as hyperthermia and hyperlocomotion, which have limited its clinical\napplication.  A study has shown that\nthe conjugation of semaglutide with MK-801 can safely enhance the\nweight-lowering properties of this NMDA receptor antagonist. The conjugation\nof MK-801 to a GLP-1 analogue was achieved through a chemically cleavable\nreducible disulfide linker. After binding to GLP-1 receptor-expressing\nneurons in the brainstem and hypothalamus, the conjugate is internalized,\nleading to the cleavage of the linker and the release of MK-801. In\nmice, the glucose-lowering effect of the MK-801-semaglutide conjugate\nwas comparable to that of semaglutide alone. At the same time, the\nMK-801-semaglutide combination resulted in an additional weight loss\nof 7%, whereas semaglutide reached a plateau. This approach demonstrates\nthe feasibility of using peptide-mediated targeting to achieve cell-specific\nmodulation of ionotropic receptors. It highlights the therapeutic\npotential of unimolecular mixed GLP-1 receptor agonism and NMDA receptor\nantagonism for safe and effective obesity treatment.\nIn 2022, nearly 20 million new cancer cases were recorded globally,\nalongside 9.7 million cancer-related deaths. Current estimates suggest\nthat approximately one in five people will develop cancer in their\nlifetime, with around one in nine men and one in 12 women succumbing\nto the disease. Lung cancer was the most commonly diagnosed form,\nfollowed by breast cancer in women, colorectal cancer, prostate cancer,\nand stomach cancer.\nTraditional\nanticancer drugs, including alkylating agents, platinum-based compounds,\nanthracyclines, topoisomerase inhibitors, and antimicrotubule agents,\noften lack specificity, targeting all rapidly dividing cells rather\nthan solely cancerous ones.  This approach\naffects normal, healthy cells as well, leading to adverse effects\nsuch as immunosuppression, hair loss, and gastrointestinal toxicity.\nAdvances have led to the development\nof monoclonal antibodies and\nantibody-drug conjugates (ADCs), which combine antibodies with cytotoxic\ndrugs or radioactive particles. These targeted therapies can focus\ndirectly on tumor cells, significantly reducing toxicity. Examples\nof monoclonal antibodies include rituximab, trastuzumab, and bevacizumab,\nwhile ADCs include brentuximab vedotin, trastuzumab emtansine, and\nsacituzumab govitecan.  Despite their\neffectiveness, these therapies face challenges related to cost, accessibility,\ndrug stability, and immune-related side effects. In addition to antibody-based\ntherapies, research has been exploring peptide-based approaches inspired\nby natural regulatory mechanisms within the body. Various peptide\ndrugs have been developed and commercialized, including somatostatin\nanalogues, which have shown promise in targeted cancer treatment.\nSomatostatin, also known as growth hormone-inhibiting hormone (GHIH),\nplays an important role as a “universal inhibitor” in\ninhibiting the secretion of various growth hormones, such as insulin,\nglucagon, gastrin, secretin, and thyroid-stimulating hormones, to\nminimize hormone fluctuations. Somatostatin was the first human peptide\nproduced using recombinant technology, paving the way for synthesizing\ncomplex peptides that were previously costly to produce synthetically\nand often induced allergic reactions when extracted from animal sources.  Following the success of recombinant somatostatin,\nGenentech and Eli Lilly pioneered the development of the first recombinant\nhuman insulin.\nSomatostatin has limited pharmacological value\ndue to its short\nlength and high instability, with a half-life of only 3 min. Systematic\nstructure–activity studies identified the FWKT peptide sequence\nwithin somatostatin, representing its β-turn pharmacophore.\nThis sequence served as a lead for developing more stable and potent\nanalogues.  Two types of somatostatin\npeptides occur naturally: somatostatin-14, a shorter variant with\n14 amino acids, and somatostatin-28, a longer form that contains the\nsomatostatin-14 sequence. Somatostatin-14 is mainly found in the central\nnervous system and pancreatic islets, where it plays a crucial role\nin inhibiting the release of growth hormone, insulin, glucagon, and\nother hormones. In contrast, somatostatin-28 is primarily located\nin the gastrointestinal tract and is released from intestinal cells\nin response to food intake; it plays a key role in regulating the\ndigestive system by inhibiting gastrointestinal hormone release and\nslowing gastric emptying. Somatostatin-14 has been the primary focus\nfor optimization and anticancer drug development.  Modifications, such as truncations, incorporation of  d -Phe at the N-terminus, and threonine alcohol at the C-terminus,\nled to the creation of the first somatostatin analogue, octreotide,\nfor treating acromegaly, breast cancer, and prostate cancer.  The enzymatic recognition site is hidden in\noctreotide, showing enhanced activity and stability with an extended\nhalf-life of 2 h.\nDespite its relatively short sequence, the\nsynthesis of octreotide\nremains challenging. One of the major difficulties is the formation\nof an intramolecular disulfide bond in the presence of a tryptophan\nresidue. Additionally, the presence of a threoninol moiety at the\nC-terminus necessitates the use of nonconventional SPPS strategies,\nparticularly in the choice of resin, linker and cleavage solution.\nAn early method developed in 1991 involves the formation of a cyclic\nacetal in solution between the two hydroxyl groups of Fmoc-threoninol\nand  p -formyl-phenoxyacetic acid. The resulting intermediate\nis then anchored to an aminomethyl resin, followed by peptide chain\nassembly on a solid phase using standard Fmoc/tBu protocols. Selective\ndeprotection of Acm-protected cysteine residues, followed by on-resin\noxidation, allows for the formation of the disulfide bridge. The peptide\nis subsequently cleaved from the resin using 20% TFA in DCM.\nAlternative strategies have since been\nexplored, including cleavage\nfrom the resin using NaBH 4 /LiBH 4  to yield the\ncorresponding alcohol under reductive conditions.  Another method employed HMP resin to synthesize the protected\nhexapeptide  d -Phe-Cys­(Acm)-Phe- d -Trp­(Boc)-Lys­(Boc)-Thr­(tBu)-Cys­(Acm),\nwhich was cleaved via aminolysis using threoninol.  However, these approaches generally afforded modest yields,\ntypically not exceeding 14%.\nImproved outcomes were obtained\nusing 2-chlorotrityl resin, which\nprovided higher yields and is also commercially available as a preloaded\nThr­(tBu)-ol-2Cl-trityl resin. Disulfide bond formation was achieved\non the fully deprotected peptide using either charcoal-catalyzed oxidation\nor an iodine solution. \n , \n  The threoninol can\nbe linked to the 2-chlorotrityl resin through an amino group. The\npeptide is then synthesized on the hydroxyl group, forming an ester\nbond. Following cleavage, an O–N shift occurs in an aqueous\nsolution, resulting in the formation of the threoninol and the linear\nform of octreotide, which is then cyclized by the formation of a disulfide\nbond ( Scheme  \n ).\na A β-amino\nalcohol is\ninitially anchored to the 2-chlorotrityl chloride (2-CTC) resin. Peptide\nelongation is then performed on the hydroxyl group, resulting in the\nformation of an ester bond. Upon cleavage from the resin, an O→N\nacyl shift occurs in aqueous solution, leading to the generation of\nthe threoninol moiety and the linear form of octreotide. Subsequently,\nthe peptide is cyclized through disulfide bond formation to yield\nthe final cyclic structure.\nA β-amino\nalcohol is\ninitially anchored to the 2-chlorotrityl chloride (2-CTC) resin. Peptide\nelongation is then performed on the hydroxyl group, resulting in the\nformation of an ester bond. Upon cleavage from the resin, an O→N\nacyl shift occurs in aqueous solution, leading to the generation of\nthe threoninol moiety and the linear form of octreotide. Subsequently,\nthe peptide is cyclized through disulfide bond formation to yield\nthe final cyclic structure.\nOther resins,\nsuch as Rink amide, have also been employed. In one\napproach, threoninol was introduced as N-Boc-O-Bzl-threoninol on a\nsuccinimidyl carbonate resin, with peptide synthesis carried out via\nthe Boc/Bzl strategy. To enable the use of the Fmoc/tBu approach,\na novel acid-labile linker was developed by condensing the two hydroxyl\ngroups of N-protected threoninol with the aldehyde group of  p -carboxybenzaldehyde. This linker can be anchored to Rink\namide resin, allowing synthesis via the Fmoc/tBu strategy. \n − \n \n  After full cleavage, disulfide bond formation was accomplished via\nair oxidation over 48 h in a dilute (ca. 1 mM) ammonium acetate/ammonium\nhydroxide buffer.\nAt the industrial level, octreotide synthesis\nhas focused on solution-phase\ncoupling of peptide fragments. The strategy disclosed in several patents\ninvolves the preparation of two tripeptides (Boc- d -Phe-Cys­(Acm)-Phe-OMe\nand Z- d -Trp-Lys­(Boc)-Thr-OMe) and one dipeptide (H-Cys­(Acm)-Thr-OMe\nor -ol), followed by methyl ester hydrolysis and fragment condensation\naccording to the [3 + 3] + 2 scheme. \n , \n  A refinement\nof this approach involved coupling the C-terminal dipeptide alcohol\nH-Cys­(Acm)-Thr-ol to a hexapeptide intermediate, Boc- d -Phe-Cys­(Acm)-Phe- d -Trp-Lys­(Boc)-Thr-OH, synthesized from a dipeptide and a tetrapeptide.\nThis strategy avoids racemization of the phenylalanine residue at\nposition 3.  Final disulfide bond formation\nwas conducted after complete deprotection using hydrogen peroxide.\nFurthermore, five distinct subtypes of somatostatin receptors (SSTR1\nto SSTR5) have been identified, with one subtype predominantly overexpressed\nin tumors. This discovery prompted the development of receptor-selective\nsomatostatin analogues, including lanreotide, vapreotide, and pasireotide.  In pasireotide, cyclization between the N-\nand C-termini extends its half-life to 12 h, making it particularly\neffective for treating Cushing’s disease compared to other\nsomatostatin analogues ( Figure  \n ,  Table  \n ).\nOn the top is the sequence of the native somatostatin peptide.\nThe blue sequence in somatostatin represents a key pharmacophore in\nsomatostatin, responsible for receptor binding. In somatostatin analogues,\nlysine (K) is conserved, while tryptophan is replaced with its stereoisomers\nin octreotide, lanreotide, vapreotide, and pasireotide. Phenylalanine\nis substituted with tyrosine, which has an additional hydroxyl group,\nand threonine is replaced with a more hydrophobic amino acid, such\nas valine, in lanreotide and vapreotide. Pasireotide incorporates\nthe highest number of unnatural amino acids. In this analogue, cyclization\noccurs between the N- and C-terminal ends, while in octreotide, lanreotide,\nand vapreotide, cyclization involves the side chains of cysteine residues,\nsimilar to native somatostatin.\nSC: subcutaneous, IV: intravenous,\nO: orally, IM: intramuscular.\nDue to their ability to bind to receptors that are\noverexpressed\nin specific tumor types, a range of somatostatin analogues have been\ndeveloped for diagnostic purposes. Diagnostic somatostatin analogues\n( Table  \n ), including\nIn-111 pentetreotide, Ga-68 DOTA-TATE, Ga-68 DOTA-TOC and Cu-64 DOTA-TATE,\nare employed for diagnostic and therapeutic purposes. \n − \n \n \n  These analogues consist of the peptide octreotide linked to a radioactive\ntracer via N-terminal chelation, which enables precise tumor detection\nand treatment. Through techniques such as peptide scintigraphy, targeted\nradiotherapy, computed tomography (CT) and positron emission tomography\n(PET), these radiolabeled peptides facilitate the visualization of\nsomatostatin receptor-expressing tumors, enhancing diagnostic accuracy\nand therapeutic efficacy.  Lu-177 DOTA-TATE\nis another approved metal-containing peptidomimetic and is used as\na therapeutic isotope.\nIV: intravenous.\nThe Ga-68-PSMA-11 complex represents another important\ndiagnostic\npeptide. Unlike the somatostatin analogues mentioned above, it is\na urea-based peptidomimetic that has become an essential tool in prostate\ncancer diagnosis. Prostate-Specific Membrane Antigen (PSMA) has emerged\nas a key biomarker and therapeutic target in oncology, particularly\nfor prostate cancer. PSMA-11 is a peptidomimetic in which the Glu-Urea-Lys\nsequence enables selective binding to PSMA.  Conjugation of PSMA-11 with the radioactive isotope Ga-68 enables\nits application in PET/CT imaging.  The\nresulting 68Ga-PSMA-11 complex offers high sensitivity and specificity\nfor the detection of metastatic or recurrent prostate cancer, playing\na crucial role in diagnosis, staging, and monitoring.  It has transformed prostate cancer imaging\nby allowing the identification of smaller lesions that may evade detection\nby conventional imaging techniques. Furthermore, 68Ga-PSMA-11 provides\nsuperior accuracy, yielding clearer and more reliable images of malignant\ntissues, and has become an integral component of modern prostate cancer\nmanagement.\nAll somatostatin analogues primarily act as agonists;\nhowever,\ncertain peptide drugs have been developed as antagonists, particularly\nthose that mimic gonadotropin-releasing hormone (GnRH) ( Table  \n ). GnRH stimulates the release\nof follicle-stimulating hormone (FSH) and luteinizing hormone (LH),\nwith LH playing a crucial role in initiating ovulation during the\nmenstrual cycle. In situations where an egg is released prematurely\nbefore it is ready for fertilization, GnRH antagonists function by\nbinding to GnRH receptors to inhibit the activity of the natural GnRH\nand preventing the premature release of the egg, thereby facilitating\nreproductive management. Notable examples of GnRH antagonists include\ncetrolix, ganirelix. \n , \n  Additionally, degarelix, and\nabarelix are effective GnRH antagonists used in the treatment of advanced\nprostate cancer ( Figure  \n ). \n ,\nAt the top is the sequence\nof the native GnRH peptide, followed\nby various commercially available analogues. Residues differing from\nthe original sequence are highlighted in orange, showcasing the extensive\nuse of unnatural amino acids. For such short sequences, numerous modifications\nare necessary to ensure a sufficiently prolonged half-life for these\npeptide drugs. Created in BioRender.\nIn these antagonists, all positions have been substituted\nwith\nD-amino acids or unnatural amino acids, except for positions 4, 7,\nand 9, which remain unchanged from the original sequence.\nConversely,\nother drugs have been developed as superagonists of\nGnRH, which serve to desensitize and downregulate GnRH receptors,\nindirectly exerting their antagonistic effects. The GnRH superagonists\ninclude leuprolide, buserelin, goserelin, nafarelin, histrelin, and\ntriptorelin. \n − \n \n \n \n \n  These agents are clinically applied in cancer treatment, puberty\nsuppression, management of estrogen-dependent female disorders, sex\nreassignment, and in vitro fertilization therapy. Superagonists have\nbeen designed by substituting the glycine at position 6 with a D-stereoisomer\namino acid. This modification enhances the peptide stability against\nproteolytic degradation and increases their receptor affinity. In\nthe native peptide, C-terminal amidation is present to confer resistance\nto carboxypeptidase degradation. In contrast, superagonists incorporate\nethylamine or a hydrazine glycine mimetic, which protects against\ncarboxypeptidase similarly to the original amidation while also increasing\nhydrophobicity (ethylamine) and conformational rigidity (hydrazine\nglycine mimetic), thereby enhancing potency and duration of action.\nUnlike GnRH antagonists, the superagonists feature fewer modifications,\nwith alterations primarily observed at positions 6 and 10.\nSeveral\nmethods have been described for the synthesis of peptide  N -alkyl amides, such as leuprolide. These methods rely on\nnucleophilic displacement of the peptide, which is anchored to a Merrifield-type\nresin, oxime resin, or polyacrylic resin. \n , \n  Standard features of these approaches include: (i) elongation of\nthe peptide chain using either Boc or Fmoc chemistry, (ii) cleavage\nof the peptide from the resin via a nucleophile, and (iii) side-chain\ndeprotection using either HF or TFA, depending on the strategy employed\nin step (i).\nIn the case of leuprolide, synthesis was performed\non a Merrifield-like\nresin using Boc chemistry. The protected peptide was cleaved from\nthe polymeric support using MeOH/TEA, followed by treatment of the\nresulting ester with ethylamine.  To\nenable the use of the Fmoc strategy, leuprolide was then synthesized\non a hydroxymethyl-Nbb resin, which can be prepared from MBHA resin\nand 4-hydroxymethyl-3-nitrobenzoic acid using alanine as an internal\nstandard.  The first four amino acids\nwere coupled using Boc chemistry to avoid the formation of diketopiperazine\n(DKP), after which the synthesis was continued using the Fmoc strategy\n( Scheme  \n ).\na To prevent DKP\nformation,\nthe first four amino acids were coupled using Boc chemistry. Following\nthis initial step, peptide chain elongation proceeded using the Fmoc\nstrategy. The final cleavage was carried out using a standard cocktail\nof TFA.\nTo prevent DKP\nformation,\nthe first four amino acids were coupled using Boc chemistry. Following\nthis initial step, peptide chain elongation proceeded using the Fmoc\nstrategy. The final cleavage was carried out using a standard cocktail\nof TFA.\nOther FDA-approved peptide-based drugs\nwith antitumoral effects\ninclude carfilzomib and romidepsin. Carfilzomib, approved in 2012,\nis indicated for treating multiple myeloma in patients who have received\nat least two prior therapies, including bortezomib and an immunomodulatory\nagent, but continue to exhibit disease progression within 60 days\nof their last treatment.  This drug is\na tetrapeptide derived from the natural products epoxomicin and eponemycin,\nwhich exhibit antitumor activity. Carfilzomib selectively inhibits\nthe chymotrypsin-like (CT-L) activity of the 20S proteasome via an\nepoxyketone moiety at the C-terminus, allowing irreversible binding\nto the CT-L site. This mechanism differentiates it from bortezomib.\nThis inhibition results in the accumulation of polyubiquitinated proteins,\nleading to cell cycle arrest, apoptosis, and suppression of tumor\ngrowth. Carfilzomib has a short half-life of approximately 30 min\nand is primarily cleared through biliary and renal excretion, a characteristic\nthat may contribute to its favorable safety profile.\nRomidepsin\nis an anticancer agent that exerts its effects through\nchromatin remodelling, specifically as a histone deacetylase (HDAC)\ninhibitor. Approved by the FDA in 2009 for treating cutaneous T-cell\nlymphoma, a rare form of non-Hodgkin lymphoma, romidepsin is a bicyclic\npentapeptide with both N-to-C terminal cyclization and a disulfide\nbond. It was originally isolated from  Chromobacterium violaceum , a Gram-negative bacterium sourced from Japanese soil.  The structure of romidepsin comprises  d -Val, DCys, Z-dehydrobutyrine,  l -Val, and (3S,4E)-3-hydroxy-7-mercapto-4-heptenoic\nacid. Romidepsin functions as a prodrug, activated intracellularly\nthrough disulfide bond reduction by glutathione. The released free\nthiols coordinate with zinc ions within the active sites of class\nI and II zinc-dependent HDAC enzymes, leading to enzyme inhibition.\nIn its active form, romidepsin is rapidly inactivated in serum, with\nan approximate half-life of 3 h. It is commercially produced through\nfermentation.\nIn addition to drugs that mimic natural hormone\npeptides, substantial\nresearch is directed toward developing peptides capable of interfering\nwith protein–protein interactions. Significant advancements\nhave been achieved, but further investigation is needed before effective\npeptide-based drugs can be developed to interfere with the tumor pathways\nthat regulate cellular proliferation. Below, we highlight some of\nthe most notable examples.\nGrb7, a crucial protein associated\nwith cancer cell proliferation\nand migration, is a noteworthy target of interest, particularly in\nbreast cancer subtypes. The main interaction of Grb7 occurs with its\nupstream signaling partners via its Src homology 2 (SH2) domain, which\nresults in Grb7 tyrosine phosphorylation and subsequent signal transduction.\nConsequently, there is a hypothesis suggesting that inhibiting the\nGrb7-SH2 domain could potentially impede breast cancer cell migration,\nalong with other signal transduction pathways associated with Grb7-SH2.\nInhibiting Grb7 has the potential to enhance the effectiveness of\nanticancer treatments. G7–18NATE, an 11-residue peptide (WFEGYDNTFPC),\ncyclized via a thioether bond from the N-terminus to the C-terminal\nthiol side chain of cysteine, exhibits robust binding to Grb7-SH2\nunder specific conditions.  Notably,\nthis binding is phosphate-dependent, with the presence of phosphate\nstabilizing the interaction and its absence resulting in reduced affinity\nand specificity. The Grb7-SH2 domain plays a critical role as an interaction\nsite, especially for peptides featuring a pYXN motif in a turn conformation,\nwhere pY represents phosphotyrosine. Researchers have explored pY\nmimetics to enhance Grb7 inhibitors, motivated by concerns about the\nimpact of the phosphate group on membrane permeability and stability.\nCarboxylic acid–based pY mimetics, such as carboxymethylphenylalanine\n(cmF) and carboxyphenylalanine (cF), have demonstrated promising binding\nto Grb7-SH2 under physiological conditions ( Figure  \n ).\nG7–18NATE\nis an 11-residue peptide cyclized through a thioether\nbond between the N-terminus and the thiol side chain of the C-terminal\ncysteine. This construct exhibits robust binding affinity for the\nGrb7-SH2 domain under specific conditions. Peptidomimetics derived\nfrom G7–18NATE, incorporating carboxymethylphenylalanine (cmF)\nor carboxyphenylalanine (cF)represented as the blue amino\nacid in the figurehave demonstrated enhanced binding to Grb7-SH2\nunder physiological conditions. Reproduced with permission from ref  . Copyright 2015, American\nChemical Society.\nThe initial success of G7–18NATE spurred\nthe development\nof a series of second-generation Grb7-SH2 inhibitors, resulting in\nanenhancement in affinity for the interaction with Grb7-SH2. These\nenhancements included the addition of a covalent tether to create\na bicyclic peptide, removal of two unnecessary amino acids at positions\n9 and 10, and incorporation of phosphotyrosine mimetics. This led\nto the creation of a nine-amino-acid bicyclic peptide scaffold named\nG7-B7, with a K D  of 0.27 μM. Despite its higher affinity\nfor Grb7-SH2 in vivo, G7-B7 exhibited lower activity than its predecessor,\nG7–18NATE, in vitro.\nSmall\nGTPases, including Ras, Rab, and Rho, play crucial roles\nin various cancer types, where their dysfunction contributes to abnormal\ncell growth and differentiation, prolonged cell survival, disturbed\nmembrane trafficking, and impaired vesicular transport. Targeting\nthe activity of these small GTPases presents an opportunity for developing\ninnovative chemotherapeutic agents in cancer treatment. A viable strategy\nto pursue this objective involves addressing the GDP-GTP exchange\nprocess in Ras, a rate-limiting step dependent on the interaction\nwith the Ras-specific guanine nucleotide exchange factor Sos. Inhibition\nof Sos-mediated Ras activation is a promising strategy for experimental\nand therapeutic intervention. Structural analyses revealed the involvement\nof multiple interactions in Ras-Sos interactions, particularly the\ninsertion of a helical hairpin from Sos into Ras switch regions. Computational\nand experimental analyses identified key residues for helix binding\nto Ras.  Stabilized helices mimicking\nthe full-length Sos αH helix were designed using the hydrogen\nbond surrogate (HBS) approach. HBS helices, preorganized and targeting\nspecific protein receptors, were chosen for their high affinity and\nspecificity. Optimization of the helical mimic sequence enhanced solubility\nand inhibitory potential against Ras-Sos association. The resulting\noptimized sequence, FEGIYRLELLKAEEAN, showed promise as a synthetic\nmimic of the Sos αH helix.\nStapled peptides have been\ninvestigated for their potential as\ncancer inhibitors, exemplified by their application in the PPI between\nthe tumor suppressor p53 and its negative regulator MDM2 and MDMX.  This interaction has garnered extensive research\nattention and has made significant progress in clinical development.\nP53, recognized as the guardian of the genome, is a crucial transcription\nfactor responsible for regulating processes such as cell cycle arrest,\napoptosis, and cellular senescence. The functional significance of\np53 is underscored by somatic mutations that deactivate p53 in up\nto 50% of human cancers. In the remaining cases, functional p53 is\noften hindered by negative regulators acting through post-translational\nmodifications or protein sequestration. Consequently, extensive efforts\nhave been directed toward overcoming these regulatory challenges and\nharnessing p53 tumor suppressor capabilities to induce cell death.\nDespite several classes of compounds reaching clinical trials, questions\npersist regarding their toxicity, off-target effects, and how to effectively\naddress mutational resistance.  To tackle\nthese challenges, researchers have explored a diverse array of peptidomimetics,\nincluding peptide hybrids, achiral peptoids, oligobenzamides, and\nfoldamers. \n − \n \n  Additionally, initiatives have been undertaken\nto create high-affinity peptide ligands through phage display technologies.  Among these innovative modalities, stapled\npeptide-based inhibitors have made significant strides in clinical\ndevelopment, serving as a platform for testing inventive staple architectures\nand chemical approaches, particularly those designed using the 1-CPS\nand 2-CPS methods to generate inhibitors for the p53-MDM2 interaction. \n − \n \n  The one-component (1-CPS) and two-component (2-CPS) thiol–ene\nreactions have proven effective in producing stapled peptides targeting\nMDM2.\nGiven\nthe considerable genetic variability observed in viruses like human\nimmunodeficiency virus 1 (HIV-1), hepatitis C virus (HCV), and SARS-CoV,\nthe rapid development of drug resistance has become a pressing concern.\nAs a result, research on antiviral peptides is relatively limited,\nwith only a few examples reported in the literature.\nStrategies\nto create antiviral drugs focus on impeding viral entry into host\ncells, a goal that can be achieved by employing peptides designed\nto mimic the binding sites of crucial proteins involved in the entry\nprocess. For example, HIV-1 relies on its viral envelope trimeric\nprotein, gp120, to initiate binding with the CD4 protein, facilitating\nits access to host T-cells. This interaction induces a structural\ntransformation in gp120, enabling the virus to subsequently attach\nto coreceptors expressed on the host cell, specifically the chemokine\nreceptors CCR5 or CXCR4.  Subsequent\nto this engagement, a structural alteration in another viral trimeric\nprotein, gp41, triggers the insertion of a fusion peptide into the\nhost cell membrane, facilitating the merging of the virus’s\nouter membrane with the cell membrane ( Figure  \n a).\n(a) HIV is an enveloped virus that requires\nfusion of its lipid\nmembrane with the target cell membrane to initiate infection. This\nprocess is mediated by glycoproteins. Specifically, the gp120 glycoprotein\nbinds to the CD4 receptor on T-cells. This interaction induces conformational\nchanges in gp120, enabling it to bind to chemokine receptors such\nas CCR5 or CXCR4. Following this, gp41 inserts its fusion peptide\ninto the target cell membrane, initiating the membrane fusion process.\n(b) Gp41 can be structurally divided into three main domains: the\nectodomain, the transmembrane (TM) region, and the cytoplasmic tail.\nThe ectodomain contains several distinct functional regions that play\ncritical roles in membrane fusion and viral infectivity. At the N-terminus,\na hydrophobic segment known as the fusion peptide (FP) is followed\nby an α-helical region termed the N-heptad repeat (NHR). A disulfide-bridged\nloop connects the NHR to a C-terminal helical region (CHR). The CHR\nis in turn linked to the TM domain by a conformationally flexible\nregion known as the membrane-proximal external region (MPER). (c)\nThe postfusion structure of the gp41 core is characterized by a six-helix\nbundle formed by the NHR and CHR regions (PDB:  1AIK ). At the center\nof this bundle is a parallel, trimeric coiled coil composed of three\nNHR helices arranged in a left-handed superhelix. Surrounding this\ncore, three CHR helices are oriented antiparallel to the NHR helices\nand wrap around the outside of the central coiled-coil trimer. Created\nin BioRender.\nInitial research focuses on exploring peptides\nthat replicate segments\nof gp41 ( Figure  \n b). Specifically, considerable attention has been directed toward\npeptides emulating the six-helix bundle configuration formed by the\nhelical structures located at the N-terminal (NHR) and C-terminal\n(CHR) regions ( Figure  \n c). The formation of this six-helix bundle in gp41 holds significant\nimportance in the fusion process between viral and cellular membranes.\nPeptides presenting segments of this six-helical bundle are believed\nto harbor the potential to disrupt its assembly, thereby hindering\nthe fusion of the virus with host cells. A study conducted by Wild\nand colleagues has elucidated the antiviral activity of a helical\npeptide mimicking the NHR, demonstrating efficacy against HIV-1.  Following this, it was noted that trimeric\nformations of the NHR-mimetic peptide displayed enhanced inhibition\nof HIV-1 entry when compared with their monomeric counterparts. The\nincorporation of covalent stabilization within these peptide trimers,\naccomplished by forming interchain disulfide bridges, led to a significant\naugmentation of their antiviral efficacy.  Similar to the NHR mimetics, peptides engineered to emulate the\nCHR of gp41 were synthesized with the objective of impeding the formation\nof the six-helix bundle.\nSubsequently,\nbuilding upon the foundation of peptides emulating\nCHR, an HIV-1 fusion inhibitor known as enfuvirtide was derived and\ngranted approval for use in 2003 ( Table  \n ,  Figure  \n ).  In addressing the\nrise and spread of enfuvirtide-resistant strains of HIV-1, a computational\nmethodology contributed to the creation of sifuvirtide as an alternative\nfusion inhibitor. Sifuvirtide exhibited notable efficacy in impeding\nthe formation of the six-helical bundle and demonstrated activity\nagainst HIV-1 variants resistant to enfuvirtide.\nAfter the insertion of the fusion peptide, gp41 undergoes\na conformational\nchange, bringing its N-terminal and C-terminal regions together to\nform a stable six-helix bundle. This structural arrangement pulls\nthe viral and cellular membranes into close proximity, facilitating\nlipid bilayer fusion. The formation of this six-helix bundle has inspired\nthe development of peptide-based inhibitors. These inhibitory peptides\nmimic specific regions of gp41, preventing the assembly of the six-helix\nbundle and thereby blocking membrane fusion and viral entry. Created\nin BioRender.\nIV: intravenous, IN: intranasal,\nSC: subcutaneous, IN: intramuscular, T: topical, O: orally.\nEnfuvirtide, commercialized as Fuzeon or T20, was\nsynthesized in\nindustry via linear SPPS using the Fmoc strategy and HBTU/HOBt as\ncoupling reagents, yielding approximately 8% of the final purified\npeptide. To improve efficiency, a hybrid SPPS/LPPS strategy was subsequently\nemployed.  In this approach, three Fmoc-protected\nfragments, each consisting of 9–16 amino acids, were synthesized\non solid phase using 2-chlorotrityl chloride (2-CTC) resin. The peptide\nfragments were then cleaved from the resin under mild acidic conditions\nand assembled in solution. Specifically, the fragment Fmoc-AA27–35-OH\nwas coupled to phenylalanine to generate Fmoc-H-AA27–36-NH 2 . Following Fmoc deprotection, the H-AA27–36-NH 2  fragment was coupled to the Fmoc-AA17–26-OH fragment.\nAfter another Fmoc removal step, the resulting H-AA17–36-NH 2  intermediate was coupled to Ac-AA1–16-OH to yield\nthe fully protected enfuvirtide precursor (Ac-AA1–36-NH 2 ) ( Figure  \n ). The final global deprotection was carried out using TFA. The individual\nfragments were obtained with purities of approximately 90%, while\nthe final deprotected peptide was isolated with a purity of about\n75%. Overall, the synthetic process afforded a yield in the range\nof 85–90%.\nEnfuvirtide is a 36-amino acid peptide synthesized in\nthree separate\nfragments that are subsequently assembled. As illustrated in the figure,\nthe three fragmentsrepresented in different colorswere\nsynthesized on 2-chlorotrityl chloride (2-CTC) resin using HBTU as\nthe coupling reagent and HOBt as a racemization suppressant. The fragments\nwere designed to remain protected and soluble in DMF or NMP and to\nminimize epimerization during solution-phase condensation, with levels\nkept below 1%. Created in BioRender.\nSebsequently, Liskamp group has developed peptides\nthat mimic the\nCD4-binding site, presenting three noncontiguous segments of the gp120\nsequence. These peptides are arranged on a molecular framework as\ncyclic loops.  This framework consists\nof triazacyclophane (TAC) scaffold. Specifically, the gp120-derived\npeptides were synthesized with cysteine residues at both the N- and\nC-termini and cyclized via a benzyl dibromide derivative. In this\napproach, the scaffold was also functionalized with azide groups,\nwhich enabled conjugation of the peptides to the TAC scaffold via\nCuAAC reaction ( Scheme  \n b).\nHowever, this specific peptide configuration\non the triazacyclophane\nscaffold did not exhibit a significant inhibitory effect on HIV-1\ninfection, indicating that alternative strategies are required to\nmore effectively mimic the CD4-binding site. In addition, the Liskamp\ngroup demonstrated that the TAC scaffold could not prevent infection\nor neutralize HCV pseudoparticles. However, their data suggested the\nefficacy of discontinuous epitope mimics as potential synthetic vaccines.\nSARS-CoV-2 and SARS-CoV entry into cells\nrelies on a crucial PPI\nbetween the spike glycoprotein (S-protein) receptor binding domain\n(RBD) of SARS-CoV and the protease domain (PD) of the human cell surface\nreceptor angiotensin-converting enzyme 2 (ACE2). Earlier studies explored\nthe use of medium-length linear peptides derived from ACE2, demonstrating\nmicromolar affinity binding to the SARS-CoV-2 S-protein RBD, degradation\nof RBD, and inhibition of ACE2-receptor mediated host cell entry of\nSARS-CoV pseudovirus in vitro.  However,\nlinear peptides often exhibit suboptimal drug-like properties, including\npoor blood plasma stability, making them unsuitable lead compounds\nfor drug discovery. In a recent investigation, efforts were made to\ncreate stable, conformationally constrained stapled analogues of the\nACE2 PD helix α1 peptide. These analogues were designed to bind\nto the receptor-binding domain (RBD) of the SARS-CoV-2 S-protein,\npreventing interaction with native ACE2 receptors. The study suggests\nthat larger ligands with enhanced binding interactions are necessary\nfor effective binding to the SARS-CoV-2 S-protein RBD. This is crucial\nto outcompete membrane-bound ACE2 and efficiently inhibit viral infection.\nPeptides have also been employed as therapeutic agents against infections,\nand several naturally derived peptides have received FDA approval.\nHowever, to date, only a limited number of peptide-based antibiotics\nhave been approved. These include gramicidin (approved in 1952) and\nbacitracin (1962), along with glycopeptides like vancomycin (1958),\nand several cyclic lipopeptides, such as polymyxin B (1964), polymyxin\nE or colistin (2016), daptomycin (2003), dalbavancin (2014), and oritavancin\n(2014) ( Table  \n ). \n − \n \n  Despite these successes, the development of new antibiotics remains\na complex task, as both small molecules and peptides are susceptible\nto bacterial resistance mechanisms. The growing problem of antimicrobial\nresistance represents a major challenge in biomedical research, emphasizing\nthe urgent need for novel antibacterial agents. For instance, resistance\nto vancomycin emerged around 1990, and to lipopeptides by 2005. \n , \n  Polymyxins, however, have retained their efficacy against highly\nresistant bacteria, although their use is limited by toxicity, especially\nto the kidneys, due to the lipid tail, and is therefore reserved for\nsevere infections in hospital settings.\nPolymyxins share a similar structural framework but differ\nat position 6 by the presence of a D-amino acidspecifically,\npolymyxin B contains  d -Phe, whereas polymyxin E (colistin)\ncontains  d -Leu. Several synthetic strategies have been developed\nto obtain polymyxins. Initial synthesis was carried out by Volger\nwho reported a method involving the preparation of peptide fragments,\nwhich were assembled in solution and then cyclized using DCC.  In contrast, Sharma later synthesized the peptide\non a solid support and performed cyclization using diphenyl phosphoryl\nazide (DPPA) in the presence of DIEA.  Other strategies involve the use of orthogonal protecting groups\nand their selective removal and replacement to enable controlled cyclization.\nFor example, ivDde was used as a temporary protecting group, which\nwas removed and replaced with Mmt; the latter was subsequently removed\nto allow for macrocyclization.  In another\napproach, Cbz protection was employed for five amine functionalities\nprior to cyclization; however, the resulting peptide exhibited poor\nsolubility in volatile solvents, complicating their removal by rotary\nevaporation after the cyclization step.\nXu et al. reported a fully solid-phase synthetic approach\nfor the\npreparation of colistin using a branched-chain strategy in which the\nside chain of diaminobutyric acid (Dab) was anchored to the resin.\nHowever, this method presents several limitations, including the time-consuming\nand costly preparation of the starting material Fmoc-Dab-OAllyl from\nFmoc-Dab­(Boc)–OH, as well as the generation of a higher number\nof impurities.  In contrast, Ramesh et\nal. demonstrated a more straightforward synthesis of a colistin analogue\nin which 6-methylheptanoic acid was substituted with decanoic acid.  Their synthesis was performed on 2-CTC resin\nusing the Fmoc/tBu strategy. An orthogonal protecting group approach\nwas employed, where all Dab side chain amines were protected as Boc,\nthreonine as tBu, and the amino group of Dab involved in the cyclization\nwas protected as Alloc. This strategy proved advantageous, as the\nuse of penta-Boc protection significantly improved the solubility\nof the peptide in common and volatile organic solvents such as DCM,\ncompared to penta-Cbz-protected analogues. Details are reported in  Scheme  \n . Peptide cleavage\nfrom the resin was performed using a minimal amount of TFA/DCM (2%),\nand the filtrate was collected over a limited quantity of DMF. The\ninclusion of DMF facilitated the safe removal of TFA; in contrast,\nusing only DCM could result in TFA accumulation during evaporation,\npotentially leading to premature deprotection of acid-labile side\nchain protecting groups. The presence of free amines at this stage\nwould compromise the subsequent cyclization step, ultimately reducing\nthe overall yield. Cyclization was conducted in solution, resulting\nin a high-yielding and convergent synthetic approach.\na Subsequent amino\nacids were\ncoupled using HBTU/DIEA in DMF, while the lipid tail was introduced\nusing DIC/HOBt in DMF and allowed to react overnight. The choice of\nDIC was based on its demonstrated stability for prolonged coupling\nreactions in SPPS. The Alloc protecting group was removed while the\npeptide remained on the resin. Peptide cleavage was performed using\n2% TFA in DCM, which enabled efficient release from the 2-CTC resin\nwithout removing the side-chain protecting groups. The partially protected\nlinear peptide was then cyclized in solution using a DMF:DCM mixture\nin a 1:50 ratio.\nSubsequent amino\nacids were\ncoupled using HBTU/DIEA in DMF, while the lipid tail was introduced\nusing DIC/HOBt in DMF and allowed to react overnight. The choice of\nDIC was based on its demonstrated stability for prolonged coupling\nreactions in SPPS. The Alloc protecting group was removed while the\npeptide remained on the resin. Peptide cleavage was performed using\n2% TFA in DCM, which enabled efficient release from the 2-CTC resin\nwithout removing the side-chain protecting groups. The partially protected\nlinear peptide was then cyclized in solution using a DMF:DCM mixture\nin a 1:50 ratio.\nDue to their associated toxicity,\npolymyxins are typically reserved\nas a last-resort treatment in hospital settings. Consequently, there\nis a growing demand for safer alternatives to combat antibiotic resistance.\nBoth Gram-positive and Gram-negative bacteria are capable of developing\nresistance; however, Gram-negative pathogens are particularly difficult\nto treat due to their impermeable outer membrane and rapid acquisition\nof resistance mechanisms. For example,  Acinetobacter baumannii , a major cause of hospital-acquired pneumonia and bloodstream infections,\nhas developed resistance to multiple antibiotics, such as carbapenems,\nhighlighting the urgent need for new therapies. In fact, for over\n50 years, no new antibiotics specifically targeting  A. baumannii  have been successfully developed.\nIn an effort to address\nthis gap, a new class of macrocyclic peptides\n(MCPs) has recently been introduced. Zampaloni et al. reported a series\nof tethered macrocyclic peptides whose mechanism of action involves\nblocking the transport of lipopolysaccharide (LPS) from the inner\nmembrane to the outer membrane in Gram-negative bacteria by inhibiting\nthe LptB 2 FGC complex.  This\nclass of peptides was identified through whole-cell phenotypic screening\nof 44,985 MCPs from Tranzyme Pharma against a panel of Gram-positive\nand Gram-negative human pathogens. A cluster of active compounds shared\na common structural motif consisting of a tripeptide subunit and a\ndiphenylsulfide tether that closed the macrocyclic ring. One compound,\nRO7036668containing an Orn-Orn-N-Me-Trp subunitdemonstrated\na minimum inhibitory concentration (MIC) of 4 mg/L against  A. baumannii  ATCC 19606. Subsequent optimization, including\nthe substitution of the central  l -Orn with  l -Lys,\ndichloro modifications on the benzene ring, and the replacement of\nthe southwestern phenyl ring with pyridine, led to the identification\nof RO7075573. This analog exhibited up to a 64-fold increase in potency\nover the initial lead. Further development yielded zosurabalpin, a\ncompound with improved pharmacokinetic properties and in vivo efficacy.\nIn mouse models of MDR and carbapenem-resistant  A. baumannii  infections, subcutaneous administration of RO7075573 provided complete\nprotection from lethal sepsis and significantly reduced bacterial\nburden in a thigh infection model. However, intravenous administration\nin rats led to toxicity, likely due to lipid precipitation in plasma.\nTo overcome this limitation, the compound was modified with a zwitterionic\ntether, resulting in zosurabalpin. This analog retained potent antibacterial\nactivity while demonstrating improved plasma stability and tolerability.\nZosurabalpin exhibited favorable physicochemical properties for clinical\ndevelopment and showed strong in vitro activity against various MDR  A. baumannii  strains. In mouse models of pneumonia, thigh\ninfection, and sepsis, zosurabalpin effectively reduced bacterial\nloads and improved survival outcomes.\nWhile antibiotic R&D\nhas seen a slowdown, two recent industry-academic\ncollaborations have identified novel antimicrobial peptide classes\ntargeting Gram-positive bacteria. Ten years after the discovery of\na complex of eight related acyldepsipeptides (ADEPs) active against  Staphylococcus  and  Streptococcus  species,\nLabischinski and colleagues reported in  Nature Medicine  the structure of the main peptide component, ADEP1, and described\noptimized synthetic variants with enhanced antibiotic properties.  Two of these optimized peptides, ADEP2 and\nADEP4, exhibited superior in vitro potency against Gram-positive bacteria\ncompared to ADEP1. In rodent models with lethal  Enterococcus\nfaecalis  infections, ADEP2 and ADEP4 matched the effectiveness\nof linezolid, a clinically used antibiotic. ADEP4 achieved an 80%\ncure rate in a sepsis model and outperformed linezolid against  Streptococcus pneumoniae  infections in rodents. The researchers\nidentified the bacterial caseinolytic protease (ClpP) as the ADEP\ntarget, demonstrating that ADEPs bind to ClpP, activate the otherwise\ninactive Clp-protease complex, and disrupt essential bacterial protein\nregulation, potentially accounting for their potent antibacterial\neffects.\nIn this context, given the urgent need to combat antibiotic\nresistance\nand the promising properties of peptides, numerous academic research\ngroups are actively investigating antimicrobial peptides as potential\ntherapeutic agents. Antimicrobial peptides (AMPs) offer a promising\nalternative foundation to fight bacteria, especially Gram-negative\nbacteria. \n , \n  AMPs are short, positively charged,\namphipathic molecules that are evolutionarily conserved across diverse\norganisms and function as natural immune effectors against pathogens.\nAMPs are among the oldest evolutionary defenses against microbial\nthreats found across the plant and animal kingdoms. Their structural\ndiversity, shaped by the unique environments of different species,\nprovides effective, rapid, and adaptive responses to pathogens. This\ndiversity has allowed AMPs to avoid becoming obsolete in the face\nof bacterial evolution, positioning them as critical agents in developing\nnew therapeutics. Several factors make resistance development against\nAMPs particularly challenging. First, AMPs typically disrupt bacterial\ncell membranes through nonspecific binding, leading to cell lysisa\nmode of action that hinders resistance ( Figure  \n ). \n , \n  Additionally, AMPs\ninterfere with bacterial cell wall and protein synthesis, providing\na dual-action mechanism that further complicates bacterial adaptation.\nAntimicrobial\npeptides (AMPs) are abundant in nature and can be\nfound in animals, plants, and microorganisms. They exhibit a variety\nof structural forms, including helical peptides (e.g., melittin),\nβ-sheet peptides (e.g., defensins), and cyclic peptides (e.g.,\npolymyxins). These peptides are typically rich in positively charged\nresidues and adopt amphipathic conformations upon folding. Due to\ntheir unique physicochemical properties, AMPs interact directly with\nbacterial lipid membranes, leading to membrane disruption. Their mechanisms\nof action include barrel-stave model (peptides insert into the membrane,\nforming transmembrane pores), toroidal pore formation (peptides induce\ncurvature in the lipid bilayer, creating a pore lined by both peptides\nand lipid head groups), micelle-like aggregation (peptides behave\nlike detergents, disrupting the membrane and forming micelle-like\nstructures), carpet model (peptides cover the membrane surface like\na carpet, disrupting the bilayer through a collective destabilization),\nand electrostatic potential alteration (the presence of positively\ncharged residues can alter the membrane potential, potentially causing\npore formation or membrane depolarization. These mechanisms disrupt\nbacterial integrity, leading to cell death, making AMPs potent agents\nagainst a broad spectrum of pathogens. Created in BioRender.\nAMPs exhibit broad-spectrum efficacy, targeting\na wide range of\npathogens, including Gram-positive and Gram-negative bacteria, fungi,\nand some viruses. Their rapid actionoften within minutescan\noverwhelm bacterial defenses before resistance mechanisms can be upregulated.\nAttempts by bacteria to alter cell membranes to evade AMP binding\nwould likely impair their viability, underscoring the potential of\nAMPs as effective therapeutic agents. Moreover, AMPs synergise well\nwith other antimicrobials, enhancing overall effectiveness and further\nreducing the likelihood of resistance. AMPs also play a role in shaping\nthe microbiome, fostering beneficial bacterial populations and discouraging\npathogenic overgrowth, which helps sustain a balanced microbial environment\nwith fewer opportunities for resistance. Given these advantages, AMPs\nrepresent a compelling starting point for the development of next-generation\nantibiotics capable of addressing both existing and emerging antibiotic-resistant\ninfections.\nDefensins and cathelicidins\n(e.g., LL-37), two mammalian AMPs,\nare particularly interesting templates for drug design due to their\neffectiveness against microbial cell membranes. Defensins, classified\ninto α-, β-, and θ-defensins, are distinguished\nby unique disulfide bridge arrangements.  They can be found in vertebrate and invertebrate animals, plants\nand fungi. Cathelicidins, found in various vertebrates, are primarily\nproduced in epithelial cells, neutrophils, and macrophages, with LL-37\nbeing the only human member of this group.  Meanwhile, other AMPs have been identified in insects, like cecropins\nand melittin, and bacteria, such as nisin and lysostaphin, though\nnone have received FDA approval to date due to the need for further\nstudies to address toxicity concerns. \n −\nAnother subset of cationic\nantimicrobial peptides featuring β-hairpin\nstructures stabilized by disulfide bridges includes protegrins, polyphemusins,\nand tachyplesin. Employing a β-hairpin mimetic strategy, Robinson\nand colleagues demonstrated heightened antimicrobial efficacy and\nprolonged plasma half-life using peptide loops similar to protegrin-1\nwere attached to the  d -Pro- l -Pro template.  The disulfide bridges were substituted with\nvarious residues, leading to the discovery of a family of template-bound\nprotegrin mimetics. Screening these mimetics identified analogues\nwith potent broad-spectrum antimicrobial activity and significantly\nreduced hemolytic effects. They exhibited direct interaction with\nthe bacterial β-barrel protein LptD in  Pseudomonas spp . This interaction involves the lipopolysaccharide transport during\nouter membrane biogenesis, distinguishing them from other antimicrobial\npeptides primarily acting through membranolytic activity. The optimal\nquantitative retention-activity relationship (QRAR) model suggested\nthat antimicrobial potency correlates with peptide charge and amphipathicity,\nwhile hemolytic effects correlate with the lipophilicity of residues\nforming the nonpolar face of the β-hairpin.\nFurthermore,\nAMPs can prevent biofilm formation and dissolve established\nbiofilms, a common cause of persistent infections.  Their versatility and ease of modification enable the design\nof innovative biomaterials, such as peptide-based hydrogels with antimicrobial\nproperties that could effectively target bacterial resistance in both\nacute and chronic infections. For example, the antimicrobial peptide\nWMR, selected for its strong antibacterial activity, is tested for\nenhanced antibiofilm effects against  Pseudomonas aeruginosa , a Gram-negative bacterium, and  Candida albicans , a pathogenic fungus. The study demonstrates how the multivalent\nmodifications of WMR peptide with short, charged sequences (GDDS and\nWKRS) on self-assembled nanostructures significantly boost antibiofilm\nproperties, providing an effective approach to tackling biofilm-associated\ninfections ( Figure  \n ).  This nanosystem offers a promising\nstrategy for designing responsive materials with heightened antibacterial\neffectiveness and potential for controlled drug release.\nMolecular\nstructures of WMR2PA, PA1, and PA2 (right) and their\nproposed self-assembled nanostructures (left) are shown. In these\nassemblies, the bioactive segment is exposed on the surface, while\nthe hydrophobic alkyl tail is buried in the core, providing the driving\nforce for self-assembly. Reproduced with permission from ref  . Copyright 2019, American\nChemical Society.\nPeptidomimetics targeting voltage-gated sodium\nchannels (VGSCs) have drawn significant attention as promising analgesics.\nResearch on pain-targeting peptides began in the 1980s, spurred by\nthe discovery of conotoxins in the venom of cone snails.  These peptides have shown potential as selective\nchemical tools for targeting ion channels and receptors. Conotoxins\nare typically composed of 10–40 amino acids, rich in disulfide\nbonds, which give them the structural stability to selectively and\npowerfully interact with ion channels, GPCRs, and transporters. To\ndate, five distinct classes of conotoxins have been identified (α-,\nδ-, κ-, μ-, ω-type), each characterized by\na unique molecular target. Among these, ω-conotoxins specifically\ninhibit Ca v 2.2 channels, also referred to as N-type voltage-gated\ncalcium channels. These channels are predominantly expressed at nerve\nterminals, dendrites, and in neuroendocrine cells, where they play\na pivotal role in neurotransmitter release and are involved in pain\ntransmission.\nZiconotide, a synthetic\nanalogue of the conotoxin peptide ω-MVIIA, is an FDA-approved\ndrug for severe and chronic pain management. This conotoxin contains\n25 amino acids and three disulfide bridges, stabilizing a small β-sheet\nthat selectively inhibits the Ca v 2.2 channel.  Approved in 2004, it gained attention for being\n1,000 times more potent than morphine without the risk of addiction.\nHowever, its drawback lies in its delivery method, requiring infusion\nvia a pump directly into the cerebrospinal fluid to reach Ca v 2.2 channels located in spinal cord neurons.  Nevertheless, the challenging physicochemical properties\nof native peptides have led to active research efforts in developing\nconotoxin peptidomimetics within both academia and the pharmaceutical\nindustry.\nZiconotide is a 25-residue peptide containing six\ncysteine residues\nthat form three disulfide bridges, which are essential for its structural\nintegrity and bioactivity. Key residues involved in the selective\ninteraction with N-type voltage-gated calcium channels include lysine\nat position 2, arginines at positions 10 and 21, leucine at position\n11, and the N-terminal amine.  A major\nchallenge in the industrial-scale synthesis of ziconotide lies in\n(a) achieving a high-yield synthesis of the linear 25-mer precursor,\nand (b) promoting efficient, native disulfide bond formation. Due\nto the presence of multiple cysteines and the need for precise disulfide\nbridge formation, orthogonal protection strategies are typically required.\nHowever, these methods are costly and often unsuitable for large-scale\nproduction.\nAttempts to promote disulfide bond formation using\nredox buffers\nsuch as oxidized and reduced glutathione (GSSG/GSH) to mimic the cellular\nfolding environment have resulted in significant formation of scrambled\nand misfolded isomers that are difficult to separate via HPLC.  As such, despite their cost, orthogonal synthesis\napproaches remain preferable for obtaining the correctly folded product.\nZhang et al. recently reported an efficient method for synthesizing\nconotoxins with three disulfide bonds using Mob, Trt, and Acm protecting\ngroups to enable regioselective disulfide formation. Their strategy\nallowed for the successful synthesis of five conotoxins with correct\ndisulfide connectivities, yielding 20–30%.\nIn efforts to develop an orally bioavailable ziconotide\nanalogue,\ncyclization has been explored as a strategy to enhance peptide stability.\nHowever, synthesizing a cyclic form of ω-conotoxin MVIIA has\nproven challenging. Backbone-cyclized analogues with fewer disulfide\nbonds have been reported, but these often lack structural integrity\nand are likely to show compromised activity.  Cyclization via native chemical ligation using a GGPG linker has\nalso been attempted, but the oxidized product was neither structurally\ncharacterized nor functionally evaluated.\nA study came from the Craik group, which employed an asparaginyl\nendopeptidase (AEP)-mediated cyclization strategy to generate backbone-cyclized\nanalogues of MVIIA.  Several AEP isoforms\nderived from plants demonstrated the endopeptidase and transpeptidase\nactivity necessary for the head-to-tail cyclization reaction, a key\nstep in the biosynthesis of cyclotides.  Linear ziconotide analogues incorporating a linker sequence were\nsynthesized via Fmoc-based SPPS, followed by cleavage, deprotection,\nand purification. Disulfide bond formation was carried out in NH 4 OAc/GnHCl buffer at pH 6.5, yielding a dominant correctly\nfolded isomer at about 25% purity ( Figure  \n ). Postfolding, the peptides were incubated\nwith AEP to generate their cyclic counterparts. Cyclic MVIIA analogues,\nincorporating six- to nine-residue linkers composed primarily of glycine\nand alanine for minimal steric hindrance, retained structural fidelity.\nMethionine at position 12 was substituted with norleucine to avoid\noxidation-related instability. These cyclic analogues inhibited voltage-gated\ncalcium channels and exhibited significantly enhanced stability in\nhuman serum and simulated intestinal fluid. This study highlights\nthe potential of AEP-mediated enzymatic cyclization as a powerful\ntool for generating structurally complex, cyclic peptide therapeuticsoffering\na viable route to improving the pharmacological properties and therapeutic\nvalue of conotoxins beyond the reach of conventional chemical synthesis.\nStrategy\nfor the synthesis of cyclic MVIIA analogues. Top: Sequence\nof ziconotide, with identical colors indicating cysteine residues\ninvolved in the same disulfide bond. Bottom: AEP-mediated cyclization\nfollowing oxidative folding. Created in BioRender.\nThe group led by Jamieson has pioneered a novel\nseries of conformationally\nconstrained peptidomimetic analogues inspired by the μ-conotoxin\nKIIIA, extracted from the venom of the marine cone snail Conus kinoshitai.  They evaluated the activity of these mimetics\nagainst human VGSCs and identified two compounds that effectively\nblocked currents in hNav1.4 and hNav1.6 channels. The primary objective\nof their investigation was to explore whether synthetic conformational\nconstraints could replace the intricate disulfide bond bridging network\nin the μ-KIIIA conotoxin peptide, resulting in more stable analogues\nthat retained bioactivity against human VGSCs. The group devised simplified\nstructures based on μ-KIIIA by substituting the complex disulfide-bonding\nnetwork with chemical staple conformational constraints. They synthesized\nseven i, i+4, and i, i+7 stapled mimetics using various chemistries,\nincluding hydrocarbon, triazole and lactam stapling, and compared\nthem to native μ-KIIIA isomers and three nonstapled control\ncompounds. Notably, only compounds featuring the i, i+7 staples demonstrated\nlow micromolar inhibition of the tested human sodium channels, Nav1.4\nfrom skeletal muscle and NaV1.6 from the CNS.\nOther analgesic\npeptides can be engineered by targeting components\nof the innate immune system, such as complement factor C5aa\npotent pro-inflammatory mediator that recruits leukocytes and activates\nphagocytic responses. CHIPS (Chemotaxis Inhibitory Protein of  Staphylococcus aureus ) is well-known for its ability to\nantagonize the C5a receptor (C5aR), thereby blocking the interaction\nbetween C5a and its receptora critical axis in complement-mediated\nimmune activation. Building on the structural framework of CHIPS,\na novel anti-inflammatory peptide named CHOPS has been synthetically\nengineered. Given the high immunogenicity associated with the full-length\nCHIPS protein, CHOPS was designed as a minimized analogue that retains\nthe essential receptor-binding residues identified from structural\nstudies of CHIPS in complex with C5aR.  CHOPS was constructed by linking two critical CHIPS-derived fragmentsresidues\nT36–L65 (N-terminal) and K95–G112 (C-terminal), both\nof which contribute to C5aR binding. These segments were joined using\na  d -Pro-Gly dipeptide linker, which promotes the formation\nof helical and β-sheet elements characteristic of the native\nCHIPS structure ( Figure  \n ). The resulting peptide is specifically tailored to engage\nC5aR while minimizing the risk of immune activation. This rationally\ndesigned analogue presents a promising therapeutic strategy for the\ntreatment of inflammatory and autoimmune disorders.\na) Cartoon representation\nof the NMR structures of CHIPS31–121\n(PDB ID:  1XEE ). The two regions interacting with the C5a receptor (C5aR) are the\nsegments 43–61 (α-helix and β1 strand) and 95–111\n(β3 and β4 strands). (b) Amino acid sequence of CHOPS.\nThe  d -Pro-Gly linker is indicated in black. c) Cartoon representation\nof CHOPS modeled based on the structure of CHIPS31–121. The  d -Pro-Gly linker is depicted in stick representation. Reproduced\nwith permission from ref  . Copyright 2010, Springer Nature under the Creative Commons\nAttribution Noncommercial License ( https://creativecommons.org/licenses/by-nc/2.0 ).\nDue to its inherent conformational flexibility,\nthe leu-enkephalin\npeptide demonstrates the capability to bind to various opioid receptors.\nWhile effective in pain relief, this peptide carries potential side\neffects such as miosis and the risk of physical dependency. Unfortunately,\nleu-enkephalin encounters challenges related to poor bioavailability\nand susceptibility to proteolytic degradation, limiting its suitability\nas a therapeutic agent. To address these issues, researchers have\nexplored macrocyclic mimetics as an alternative approach. Blomberg\nand collaborators have detailed the incorporation of a β-turn\nmimetic, encompassing both 10- and 7-membered rings, to replace the\ninitial four residues of leu-enkephalin.  The 7-membered ring analogue lacks one glycine in the sequence,\nwhile the 10-atom cycle adopts a β-turn conformation. In both\nmimetics, the intramolecular hydrogen bond has been replaced with\nan ethylene bridge, and the amide bond between Tyr1 and Gly2 has been\nsubstituted with an isostere composed of methylene ether ( Figure  \n ). Characterization\nof these analogues was compared to their respective linear counterparts.\nThis study has unveiled that all analogues, with the exception of\nthe β-turn mimetic, can effectively interact with opioid receptors,\nproviding insights into the mechanism of action of this peptide.\nA peptidomimetic\n(center) designed with a covalently bonded 10-membered\nring that mimics the β-turn observed in crystalline leu-enkephalin\n(left). A second peptidomimetic (right) features a 7-membered ring,\ninducing a different turn conformation compared to leu-enkephalin.\nAtoms highlighted in red are conserved across all structures, while\natoms in blue represent the original segment from leu-enkephalin and\nits corresponding portion in the second peptidomimetic.\nOther FDA-approved peptides that act on the nervous\nsystem include\ndifelikefalin and trofinetide. \n , \n  Unlike conotoxins,\nthese peptides have simpler, shorter, and linear structures, lacking\ncyclic components. They are used for the treatment of itch and Rett\nsyndrome, respectively ( Table  \n ).\nIV: intravenous, IN: intramuscular,\nO: orally.\n\nOral administration remains the most patient-friendly route of\ndrug delivery. However, achieving effective bioavailability for large,\nhydrophilic peptides remains a formidable challenge. Along their journey\nfrom ingestion to absorption, peptides encounter multiple physiological\nbarriers that significantly hinder their therapeutic potential ( Figure  \n ). One of the primary\nobstacles is the harsh gastric environment. Peptides are generally\nunstable in the acidic pH of the stomach and are susceptible to enzymatic\ndegradation by pepsins, rendering them unsuitable for oral delivery.\nFurther down the gastrointestinal tract, peptides face proteolysis\nby intestinal enzymes, including endopeptidases such as trypsin, chymotrypsin,\nand elastase, as well as exopeptidases like aminopeptidase N, dipeptidases,\nand carboxypeptidases A and B. In addition, lysosomal enzymes within\nintestinal epithelial cells (enterocytes) further contribute to peptide\ndegradation.\nIllustration of the\nmain barriers faced by peptides during oral\nabsorption. After oral administration, peptides first encounter the\nacidic environment of the stomach, where the pH is low, leading to\npotential degradation. As they progress toward the intestinal epithelium,\nthe pH becomes more basic. Peptides must then penetrate the mucus\nlayer that protects the underlying intestinal epithelial cells (enterocytes)\nand avoid degradation by various digestive enzymes. Upon reaching\nthe enterocytes, peptides can cross the epithelial barrier either\nvia transcellular transport (passing through the cells) or via paracellular\ntransport (passing between the cells through transient openings in\nthe tight junctions). Created in BioRender.\nBeyond enzymatic breakdown, the intestinal mucus\nlayer poses another\nsignificant barrier. While its mesh-like structure features pores\nlarge enough to permit peptide diffusion, interactions between peptides\nand mucus componentsparticularly hydrophobic interactions\nand hydrogen bondingcan severely impede their diffusion.  Furthermore, peptides containing thiol or disulfide\ngroups are especially vulnerable to thiol–disulfide exchange\nreactions, which can inactivate the drug. These exchanges may occur\nwith glutathione, cysteine-rich mucus glycoproteins, or dietary proteins\ncontaining free cysteine residues.  Following\nthese barriers, peptides must still traverse the intestinal epithelium\nto reach systemic circulation. This can occur via two primary routes:\nthe transcellular pathway, which involves crossing the apical membrane,\nmigrating through the cytoplasm, and exiting via the basolateral membrane;\nor the paracellular route, which requires passage through tight junctions\nbetween adjacent epithelial cells. The latter is highly restrictive,\npermitting only small and transient openings that limit paracellular\ntransport of peptide drugs.\nVarious\napproaches can enhance peptide transport and intestinal\nabsorption. Modifying peptide sequences to make them less susceptible\nto protease degradation is effective; common strategies include using\nnonproteinogenic amino acids or cyclizing the peptide ( Section  \n ). However, these modifications\nare often insufficient and require additional protective systems.\nLipid- or polymer-based carriers are among the most effective systems,\nserving as protective shields and transport vehicles for peptides. \n , \n  Additionally, an enteric coating can help improve peptide absorption\nthrough the gastrointestinal (GI) tract. Enteric coating is a polymeric\nbarrier, such as methacrylic acid copolymers and hydroxypropyl methylcellulose\nphthalate, applied onto the surface of the oral drug with the aim\nof protecting it from the acidity of the stomach and release the drug\nin the upper tract of the intestine. \n ,\nOne\nof the most widely studied strategies is the use of permeation\nenhancers in peptide administration. These enhancers are typically\nnonionic surfactants chosen for their low toxicity and minimal reactivity.\nWhen included in drug formulations, surfactants can enhance peptide\npermeation by integrating into the cell membrane, disrupting the structural\nintegrity of the lipid bilayer. This disruption compromises the membrane’s\nbarrier function, increasing permeability and fluidity.  Several surfactants, including sodium dodecyl\nsulfate, sodium taurodihydrofusidate, polyoxyethylene ethers, and\nmedium- to long-chain fatty acids, such as capric acid (decanoic acid)\nand caprylic acid (octanoic acid), have been used in oral drug formulations,\noften in combination with other carriers.\nDespite their therapeutic\npotential, there are currently few examples\nof peptides successfully administered via the oral route; the majority\nare still delivered through injections. However, extensive research\nis ongoing to overcome the challenges associated with oral peptide\ndelivery, and several candidates are currently undergoing clinical\nevaluation. For instance, novel oral formulations of peptides such\nas insulin (ORMD-0801), calcitonin (SMC021), and difelikefalinadministered\nin combination with permeation enhancershave progressed to\nclinical trials, with some reaching Phase 3. \n − \n \n  Additionally, oral formulations of leuprolide have advanced to Phase\n2 trials.\nNovo Nordisk developed\nan oral formulation of semaglutide, marketed\nas Rybelsus, which is administered once daily. This formulation utilizes\nthe Eligen technology created by Emisphere Technologies, where semaglutide\nis coformulated with sodium N-[8-(2-hydroxybenzoyl) amino] caprylate\n(SNAC). SNAC acts as an absorption enhancer, facilitating gastrointestinal\nuptake ( Figure  \n ).  SNAC functions by forming a noncovalent complex\nwith semaglutide, thereby increasing its lipophilicity and enabling\ntranscellular transport across the gastrointestinal epithelium. The\nSNAC-based formulation used for Rybelsus also includes additional\nabsorption enhancers such as N-(5-chlorosalicyloyl)-8-aminocaprylic\nacid (5-CNAC), 4-([4-chloro-2-hydroxybenzoyl]-amino) butanoic acid\n(4-CNAB), and N-(10-[2-hydroxybenzoyl]-amino) decanoic acid (SNAD).\nThese components form a complex with the peptide that remains insoluble\nat low pH, thereby protecting it from degradation by gastric peptidases.\nUpon reaching the small intestine, where the pH exceeds 7, the complex\ndissociates, allowing the peptide to be absorbed efficiently.\nMoreover, the absorption specificity\ndepends on the properties\nof the therapeutics used. For instance, liraglutide is not absorbed\nwhen coformulated with SNAC, likely due to its higher hydrophobicity.\nSimilarly, using a closely related analogue of SNAC does not facilitate\nthe absorption of semaglutide, underscoring the importance of the\nspecific interaction between semaglutide and SNAC for effective oral.  These findings underscore the challenges of\ntranslating absorption-enhancing technologies from one drug candidate\nto another, even within the same therapeutic class.\nAstraZeneca\nhas also developed an alternative formulation for oral\nadministration of a peptide-based antdiabetic therapeutic, which has\nbeen directly compared to oral semaglutide in terms of pharmacokinetics,\nbioavailability, and clinical efficacy. They modified the native GLP-1\npeptide by incorporating multiple α-methyl amino acids at specific\npositions vulnerable to proteolytic attack, resulting in the analogue\nJ211. To extend its circulating half-life, lipidation was performed\nat position 26, similar to the modification strategy used for semaglutide,\nwhere lysine at position 26 was conjugated with a linker and a C 18  dicarboxylic lipid. For enhanced potency, J211 underwent\na lipidation scan to identify optimal sites for lipid attachment.  As a result, positions 19 and 31 were substituted\nwith lysine residues, which were further functionalized with dodecanoic\nacid, producing MEDI7219, the first bis-lipidated GLP-1 analogue.\nMEDI7219 was formulated as enteric-coated oral tablets containing\n100 mg of sodium chenodeoxycholate (NaCDC) and 200 mg of propyl gallate\n(PG) as permeation enhancers. The enteric coating was designed to\nprotect the formulation from the acidic environment of the stomach\nand ensure drug release in the neutral pH of the intestine. For comparison,\nsemaglutide tablets were formulated without enteric coating, incorporating\n20 mg of the peptide and 300 mg of SNAC as the permeation enhancer.\nIn pharmacokinetic studies, the oral bioavailability of MEDI7219 in\ndogs was significantly higher than that of semaglutide (5.92% vs 0.08%).\nHowever, MEDI7219 exhibited a shorter plasma half-life compared to\nsemaglutide (9.8 h vs 60.5 h), consistent with the lower plasma protein\nbinding observed in vitro. These pharmacokinetic parameters suggest\nthat MEDI7219 is suitable for once-daily oral dosing in its current\ntablet formulation.\nSeveral studies have investigated the use\nof protease inhibitors\nto improve the bioavailability of orally administered proteins. For\nexample, the coadministration of calcitonin with aprotinin reduced\ncalcitonin degradation in the colon but did not increase its plasma\nconcentration.  Small-molecule inhibitors\nsuch as camostat mesylate, bacitracin, soybean trypsin inhibitor,\nand aprotinin have also been tested for their effects on insulin metabolism.\nWhile camostat mesylate and bacitracin improved insulin bioavailability\nin the large intestine, they did not affect absorption in the small\nintestine.  Rapid dilution, low potency,\nand digestion-related issues can limit the effectiveness of these\ninhibitors. Higher doses could overcome these limitations but raise\nsafety concerns, including pancreatic hypertrophy, hyperplasia, and\nnephrotoxicity. Additionally, the pancreas may counteract inhibitor\neffects by increasing protease secretion, and these inhibitors may\nalso disrupt the absorption of other proteins, affecting overall gastrointestinal\nmetabolism.\nAdditionally, alternative delivery routes for GLP-1\nreceptor agonists\nhave been explored. MannKind Corporation developed an inhalable GLP-1\npowder, MKC253, using the Technosphere platform. In this method, GLP-1\nis adsorbed onto fumaryl diketopiperazine (FDKP) microparticles with\na size range of 2–5 μm. Upon reaching the lungs, FDKP\ndissolves, allowing GLP-1 to be absorbed into the systemic circulation.  Although early Phase 1 trials showed potential\nbenefits, such as improved systemic delivery and reduced gastrointestinal\nside effects, further development was discontinued due to strategic\nchallenges and inconsistent therapeutic effects.\nIn addition\nto all the strategies mentioned above, macroscopic\nmaterials, often classified as medical devices, have been widely utilized\nin oral drug delivery systems. Some of these systems incorporate combinations\nof these materials, including osmotic capsules and microneedles. For\ninstance, the osmotic-controlled release oral delivery system (OROS),\nan FDA-approved technology, is designed to provide controlled, extended\nrelease of drugs over time. It consists of a rigid capsule containing\na core with the active pharmaceutical ingredient and a semipermeable\nmembrane that governs the drug release rate.  Upon contact with the gastrointestinal tract, the OROS capsule absorbs\nwater, causing the core to expand. The osmotic pressure generated\ngradually releases the drug through a small hole in the membrane.\nThis technology is particularly advantageous for drugs that require\nstable plasma concentrations over prolonged periods, reducing the\nfrequency of dosing and minimizing fluctuations in drug levels. It\nis unaffected by variables such as pH, food intake, and intestinal\nenvironment. Drugs delivered through OROS include extended-release\nformulations like Concerta (methylphenidate for ADHD) and Cardura\nXL (doxazosin for hypertension). However, OROS has limitations, such\nas the complexity of manufacturing and potential gastrointestinal\nirritation or even blockage of the gastrointestinal tract due to the\nprolonged release of certain drugs.\nMore recently, innovative\ndrug delivery technologies, such as coated\nor integrated microneedles, have been developed for peptide delivery.\nOne notable example is the RaniPill, developed by Rani Therapeutics. \n , \n  This capsule sheds its cellulose coating upon reaching the intestine,\ntriggering the inflation of a balloon inside the capsule. This inflation\ncreates enough pressure to push the microneedles out of the capsule,\nallowing them to penetrate the intestinal wall and deliver the drug\ndirectly into the bloodstream. This technology has demonstrated over\n50% oral bioavailability in preclinical studies for insulin and adalimumab.  However, its utility is limited by the relatively\nsmall drug payload (3–5 mg per pill). Although the first-in-human\nsafety study reported no adverse events, the small payload restricts\nthe broader application of this technology.\nOther emerging microneedle\ntechnologies, including self-orienting\nmillimeter-scale applicators (SOMA), are in development ( Figure  \n ). SOMA device\nis designed to deliver its drug payload to the stomach lining via\na fluid-triggered dissolution process that deploys a spring mechanism\nto inject the drug.  Although initial\nstudies have shown promise, with SOMA capable of delivering up to\n0.5 mg of insulin per device, advancements have led to a new version\ncapable of delivering doses up to 4 mg. This updated SOMA device demonstrates\nup to 80% bioavailability within hours of administration, making it\na promising candidate for the delivery of both small molecules and\nmonoclonal antibodies.  Additional optimization\nof the device is required to reduce the capsule size, increase drug\nloading capacity, and further minimize the risk of gastrointestinal\nobstruction.\nTop: Structure of SNAC, an absorption enhancer by locally\nincreasing\npH and promoting transcellular transport across the gastric epithelium.\nBottom: Schematic representation of emerging microneedle technologies\nSOMA. The SOMA device is engineered to deliver drug payloads directly\nto the stomach lining through a fluid-triggered dissolution process\nthat activates a spring-loaded injection mechanism. Created in BioRender.\nA widely adopted strategy to enhance peptide permeability\ninvolves\nstructural modification, most notably cyclization. Cyclization confers\nincreased proteolytic stability, protecting peptides from enzymatic\ndegradation and thereby facilitating improved intestinal absorption.\nPye and colleagues conducted a study to investigate the effects of\nmolecular size and lipophilicity on membrane permeability, utilizing\nlibraries of cyclic peptides ranging from octapeptides to decapeptides,\nwith molecular weights (MWs) between 800 and 1200 Da.  To minimize the influence of intramolecular\nhydrogen bonding on conformational preferences and membrane permeability,\nthe researchers fully  N -methylated the backbone amide\nbonds. Each peptide was designed to include one tyrosine (Tyr) and\none proline (Pro) residue, with the remaining residues limited to\namino acids featuring either natural or non-natural aliphatic side\nchains. This design constraint reduced the impact of polar and charged\ngroups, enabling a focused examination of how molecular size affects\npermeability. For assessing membrane permeability, the team employed\nthe parallel artificial membrane permeability assay (PAMPA) and utilized\nan MDCK cell clone that expressed low levels of P-glycoprotein to\nminimize transporter-mediated efflux effects. This study found a significant\ndecrease in passive permeability for peptides exceeding a molecular\nsize threshold of approximately 1000 Da. This finding suggests a fundamental\nlimitation in cellular permeability for larger molecules, challenging\ntraditional solubility-diffusion theories and proposing a potential\nmechanism involving diffusion through polymer networks.\nAdditionally,\nthe research underscored the delicate interplay between\nlipophilicity and size in achieving optimal cell permeability and\naqueous solubility, particularly within the challenging MW range of\n700–1000 Da. These observations correlate with previous data\nindicating that few orally administered drugs and clinical candidates\nexceed MWs of 1000 Da. Overall, the findings extend beyond cyclic\npeptides to include various classes of cell-permeable and orally administered\ndrugs. They highlight the importance of molecular flexibility in adapting\nto physiological conditions, thereby integrating aqueous solubility,\ncell permeability, and efficient target binding. Incorporating this\nadaptable behavior into drug design may facilitate the discovery of\nlarger drugs that expand the boundaries of cell-permeable drug space.\nIn the early 2000s, a surge of companies began exploring the potential\nof constrained peptides, motivated by the belief that these molecules\ncould target previously inaccessible intracellular sites. Technologies\nsurrounding constrained peptides, developed by various biotech firms,\nare increasingly attracting interest from larger pharmaceutical companies.\nFor instance, PeptiDream has transferred its technology platform to\nmajor pharmaceutical companies, such as Merck, Lilly, Bristol-Myers\nSquibb, Novartis, Genentech and Astellas, to enhance their drug discovery\ninitiatives.  This platform combines\nadvanced methods to generate macrocyclic peptides and screen them\nfor potential drug candidates. Developed from RaPID, the groundbreaking\nwork of the Suga team in Japan, the platform enables the rapid and\nefficient identification of novel compounds targeting specific proteins.\nIn addition to utilizing the 20 standard amino acids, the technology\nincorporates more than 3,000 nonstandard amino acids into macrocyclic\npeptides. This allows for the creation of libraries containing trillions\nof structurally diverse peptides, providing exceptional flexibility\nfor various applications.\nInnovative\nstartups continue to push the boundaries in this field.\nFor example, FogPharma is developing next-generation stapled peptides\nas miniproteins. Their efforts focus on advancing simple macrocycles,\nstapled peptides, and peptides with multiple loops, all designed to\nmimic critical binding epitopes of proteins, including β-hairpins\nand α-helices. Such capabilities allow these constrained peptides\nto disrupt targets that are often difficult for existing small molecules\nor biological therapies to affect.\nDeveloping strategies for efficient cellular peptide delivery and\ncreating orally bioavailable peptide therapeutics proved to be complex.\nChallenges related to pharmacokinetics, manufacturing, and immunogenicityparticularly\nwhen peptide lengths exceed approximately 15 amino acidshave\nintroduced further barriers, impeding progress in the field. \n ,\n\nAlthough\nthe first peptide-based drug, insulin, was discovered\nin 1921, it took several decades for the industrial development of\npeptide drugs to gain real momentum. The past three decades have witnessed\npeptides taking center stage, particularly in the treatment of diabetes\nand cancer. Numerous advancements have been made in developing synthetic\nmethodologies capable of producing peptides with high purity and efficacy.\nPeptide therapeutics occupy an intermediate position between small\nmolecules and biologics, requiring specialized expertise and tailored\napproaches for their synthesis and purification. While peptides share\ncertain attributes with proteins, their production demands distinctly\ndifferent methodologies, fostering the growth of a specialized branch\nof medicinal chemistry focused on peptide discovery and optimization.\nThis field provides tools for refining peptide structures and pharmacological\nproperties. Nevertheless, achieving an ideal peptide drug with simplified\nadministration remains an ongoing challenge, and innovations in both\ndelivery and stabilization are critical to broadening peptide therapeutic\napplications.\nToday, the field has moved well beyond merely\nreproducing natural\npeptides; peptide engineering allows the creation of novel, improved,\nand more effective peptides, thanks to continuous innovations in both\nchemical synthesis and molecular design. This shift marks a pivotal\nevolution from natural mimicry to true molecular innovation. These\ninclude the development of new protecting groups, coupling reagents\nand hybridization of SPPS and LPPS, enabling the synthesis of increasingly\ncomplex peptides incorporating non-natural amino acids.\nWhile\nnew synthetic methodologies were being developed, modern\nbiotechnological techniques based on genetic engineering were also\nintroduced. These approaches now complement and enhance chemical methods,\nenabling more efficient and versatile peptide production. In particular,\ngenetic engineering has yielded excellent results. Among its advantages\nare scalability, allowing continuous production of peptidomimetics\nwithout the need for costly chemical reagents or labor-intensive synthesis\nsteps; the use of expanded genetic codes, allowing the incorporation\nof non-natural amino acids; fewer synthesis steps, enabling the production\nof complex peptidomimetics with greater automation and fewer manual\ninterventions; high reproducibility once a genetic construct is optimized;\nand the ability to introduce post-translational modifications (e.g.,\nphosphorylation, glycosylation) that add complexity and functionality\nsometimes unattainable through synthetic methods.\nHowever, despite\nthese advances, genetic engineering is not without\nlimitations. While it offers the advantage of scalable peptide production\nat a relatively low cost for less complex peptides, challenges remain\nwhen it comes to optimizing expression systems for highly complex\nor hydrophobic peptides. These peptides tend to aggregate, becoming\ninsoluble, which makes them difficult to express and purify effectively,\nespecially at industrial scales. Purification processes can still\nbe costly and labor-intensive, and the need for specialized reagents\ncan drive up expenses. On the other hand, classic synthetic chemistry\nexcels in providing precise control over peptide design. It allows\nthe incorporation of non-natural amino acids, giving researchers complete\nfreedom to tailor the sequence, composition, and stereochemistry without\nbeing constrained by the limitations of the genetic code. This flexibility\nis a key strength of synthetic chemistry, but it does come with its\nown set of challenges. Synthetic peptide production can be labor-intensive,\ntime-consuming, and expensive, particularly when scaling up for industrial\nproduction of large peptides.\nBoth genetic engineering and synthetic\nchemistry have their distinct\nadvantages, and neither approach is universally superior. Genetic\nengineering shines in scalable, cost-effective peptide production,\nwhile synthetic chemistry remains unparalleled in precision and the\nincorporation of non-natural elements. The choice of method often\ndepends on specific project requirements and the balance between cost,\nscalability, and customization.\nAn additional advantage of biotechnological\napproaches lies not\nonly in peptide production but also in drug screening. Systems like\nRaPID exemplify this capability, allowing for the rapid discovery\nof effective sequences against specific diseases. Biotechnologies\ncontribute not only to production and drug discovery, but offer methods\nthat are inherently more sustainable and “green” compared\nto purely chemical synthesis.\nDetermining whether synthetic\nor biotechnological approaches are\n“better” is complex. Each method complements the other,\nand hybrid approaches combining chemical synthesis with genetic engineering\nmay offer the best of both worlds, leveraging the strengths of each\ntechnique. For instance, synthetic modifications could be introduced\nafter expression, merging the biological efficiency of genetic engineering\nwith the chemical versatility of synthesis. Alternatively, peptides\ncould be synthesized in fragments and then assembled using enzymatic\nmethods, similar to the process used in CEPS.\nThe past decade\nhas witnessed significant successes in peptide\ntherapeutics, notably glucagon-like peptide-1 (GLP-1) analogues, which\nhave revolutionized diabetes and obesity management. Among these,\nsemaglutide (marketed as Ozempic) emerged as the top-selling GLP-1\nagonist in 2023, with 2024 sales projections exceeding $16 billion.\nCombined forecasts for semaglutide-based therapies, including Rybelsus\n(oral semaglutide) and Wegovy (for obesity), are expected to surpass\n$28 billion in 2024. Beyond diabetes, peptides have shown promise\nacross diverse therapeutic areas, including pain management, infectious\ndiseases, oncology, and diagnostics. Despite these advances, peptide\ntherapeutics still face significant barriers to entry in some fields,\nnotably neurology.\nCurrently, no peptide-based therapeutics\nhave been approved for\nneurological diseases, largely due to the challenge of crossing the\nblood-brain barrier (BBB). Antibodies have demonstrated greater success\nin this domain, owing to engineered transport mechanisms. No broadly\nreliable method yet exists for consistent peptide delivery across\nthe BBB. Emerging strategies, such as peptides derived from viral\nproteins, offer promising solutions by leveraging natural mechanisms\nof BBB penetration. Innovations in receptor-binding neuropeptides\ncould provide cost-effective alternatives to antibody-based therapies.\nAnother promising avenue involves targeting the gut-brain axis, though\nthis introduces the additional challenge of overcoming gastrointestinal\nbarriers. While significant research effort is directed toward these\nsolutions, the clinical translation of such strategies remains in\nits infancy.\nAntibodies have established a dominant position\nacross therapeutic\nareas such as oncology, autoimmune diseases, and infectious diseases,\ndue to their ability to precisely target specific proteins or cells.\nNotable examples include pembrolizumab (Keytruda) and adalimumab (Humira),\nboth among the top-selling antibody therapeutics in 2023. Although\npeptides are gradually expanding their clinical footprint, they are\nunlikely to replace antibodies in areas where long systemic half-life\nand structural robustness are critical. Instead, peptides will find\nniche applications by targeting intracellular pathways inaccessible\nto antibodies, offering a complementary rather than competitive therapeutic\napproach.\nDelivery remains a pivotal challenge in peptide drug\ndevelopment.\nMost peptide therapeutics are administered via injectiona\nroute that, while effective, is less preferred by patients. Oral delivery\npresents significant hurdles, as peptides must withstand the harsh\ngastrointestinal environment. Achieving oral bioavailability often\nrequires much higher doses, increasing production costs substantially.\nWhile cyclic peptides offer greater stability for oral delivery, their\nbioavailability remains limited. Despite these challenges, the success\nof injectable peptides in improving outcomes for diabetes and cancer\nhas redefined the concept of the “ideal” drug: oral\nbioavailability is desirable, but no longer an absolute requirement.\nMembrane permeability and structural stability continue to be major\nobstacles. Strategies such as conjugation with cell-penetrating peptides\n(CPPs) and prodrug development have shown promise, but no broadly\napplicable solution has yet emerged.\nDespite remarkable progress,\npeptide-based drugs continue to face\nsignificant challenges limiting their widespread application. Advances\nin chemical modifications, formulation technologies, and delivery\nsystems have substantially improved peptide stability, bioavailability,\nand pharmacokinetics. Nevertheless, peptides remain susceptible to\nenzymatic degradation, short systemic half-lives, immunogenicity,\noff-target effects, and manufacturing complexity.\nFuture research\ndirections will likely emphasize hybrid approaches\ncombining the precision of synthetic chemistry with the scalability\nof biotechnology. Novel delivery platforms, such as nanoparticle-based\ncarriers, viral vector systems, and next-generation oral formulations,\nwill be critical to expanding the clinical applications of peptides.\nMoreover, deeper exploration of intracellular targets, modulation\nof the gut-brain axis, and improvements in BBB-penetrating technologies\ncould open entirely new therapeutic landscapes.\nIn the long\nterm, peptide therapeutics are poised to complement,\nrather than replace, antibodies and small molecules, carving out their\nown essential role within the increasingly sophisticated toolbox of\nmodern medicine. Their success will depend on striking the right balance\nbetween biological complexity, therapeutic efficacy, patient convenience,\nand manufacturing feasibilitya formidable but exciting scientific\nfrontier.","source_license":"CC-BY-4.0","license_restricted":false}