# Peptides — Map of Content --- ## What Peptides Are Peptides are short chains of amino acids linked by amide (peptide) bonds. Convention draws the line at roughly 50 residues: below that threshold, a molecule is a peptide; above it, a protein. The distinction matters because peptides fold differently, clear the body faster, and penetrate tissues that proteins cannot reach. Every peptide begins with an amino terminus (N-terminus) and ends with a carboxyl terminus (C-terminus). The sequence of residues between those ends — the primary structure — dictates the molecule's shape, binding affinity, and biological function. --- ## Classification by Origin ### Endogenous Peptides (Produced by the Body) - **Insulin** (51 residues) — regulates glucose uptake; produced by pancreatic β-cells. Two disulfide-linked chains (A and B) control blood sugar within a narrow range. - **Oxytocin** (9 residues) — drives uterine contraction during labor and mediates social bonding. A cyclic nonapeptide with a disulfide bridge between Cys-1 and Cys-6. - **Vasopressin (ADH)** (9 residues) — controls water reabsorption in the kidneys. Differs from oxytocin by just two amino acids, yet triggers entirely different receptor cascades. - **Glucagon** (29 residues) — raises blood glucose by stimulating hepatic glycogenolysis. Acts as insulin's counter-regulatory partner. - **Bradykinin** (9 residues) — dilates blood vessels and lowers blood pressure. Released from kininogen by the enzyme kallikrein. - **Endorphins** (β-endorphin: 31 residues) — bind μ-opioid receptors to suppress pain and produce euphoria. The body's endogenous analgesic system. - **Angiotensin II** (8 residues) — constricts arterioles and stimulates aldosterone release. Central to the renin-angiotensin-aldosterone system that governs blood pressure. - **Atrial Natriuretic Peptide (ANP)** (28 residues) — opposes angiotensin II by promoting sodium excretion and vasodilation. Secreted by atrial cardiomyocytes in response to stretch. - **Ghrelin** (28 residues) — the "hunger hormone." Secreted by gastric cells, it activates the growth hormone secretagogue receptor (GHS-R1a) in the hypothalamus. - **Substance P** (11 residues) — a neuropeptide of the tachykinin family. Transmits pain signals and drives neurogenic inflammation through NK1 receptors. ### Exogenous Peptides (From External Sources) - **Antimicrobial peptides (AMPs)** — magainins from frog skin, defensins from insects. These host-defense molecules disrupt bacterial membranes by forming pores or carpet-like coatings. - **Conotoxins** — cone snail venoms that block specific ion channel subtypes (Na⁺, K⁺, Ca²⁺) with extraordinary selectivity. Ziconotide (ω-conotoxin MVIIA) treats severe chronic pain. - **Melittin** (26 residues) — the principal toxin in bee venom. An amphipathic α-helix that lyses cell membranes. Under investigation as an anti-cancer and anti-microbial agent. - **Cyclosporine** (11 residues, cyclic) — a fungal peptide that suppresses T-cell activation by inhibiting calcineurin. Transformed organ transplantation. --- ## Synthetic Peptides ### Solid-Phase Peptide Synthesis (SPPS) Bruce Merrifield invented SPPS in 1963 and earned the Nobel Prize for it in 1984. The method anchors the first amino acid to an insoluble resin bead, then adds residues one at a time from the C-terminus to the N-terminus. Each cycle involves three steps: deprotection of the α-amino group, coupling of the next protected amino acid, and washing to remove excess reagents. Two protection strategies dominate: - **Fmoc (9-fluorenylmethoxycarbonyl)** — removed by piperidine (base-labile). The current standard. Milder conditions preserve sensitive side chains. - **Boc (tert-butyloxycarbonyl)** — removed by trifluoroacetic acid (acid-labile). Merrifield's original approach. Still used for difficult sequences. SPPS routinely produces peptides up to 50 residues. Beyond that length, coupling efficiencies compound into significant yield losses. ### Native Chemical Ligation (NCL) Stephen Kent introduced NCL in 1994 to overcome the length barrier. The method joins two unprotected peptide fragments: one bearing a C-terminal thioester, the other an N-terminal cysteine. A transthioesterification followed by an S→N acyl shift forms a native peptide bond at the ligation site. NCL extended chemical synthesis to proteins exceeding 200 residues. ### Recombinant Expression For large-scale production, engineered bacteria (*E. coli*), yeast (*Pichia pastoris*), or mammalian cells (CHO) express peptides from cloned genes. Insulin manufacturing shifted from porcine extraction to recombinant *E. coli* production in 1982 — the first recombinant pharmaceutical approved by the FDA. ### Key Synthetic Peptide Examples - **Semaglutide** — a GLP-1 receptor agonist with a C-18 fatty diacid that binds albumin, extending half-life to ~7 days. Treats type 2 diabetes (Ozempic) and obesity (Wegovy). The acylation strategy — attaching a lipid moiety to Lys-34 via a linker — represents a breakthrough in peptide pharmacokinetics. - **Liraglutide** — the predecessor to semaglutide. A C-16 fatty acid conjugate of GLP-1 with a half-life of ~13 hours. Proved that lipidation could rescue peptides from rapid renal clearance. - **Leuprolide (Lupron)** — a synthetic GnRH agonist that downregulates gonadotropin release through receptor desensitization. Treats prostate cancer, endometriosis, and precocious puberty. - **Octreotide (Sandostatin)** — a synthetic somatostatin analogue. Eight residues mimic the activity of the 14-residue natural hormone, with a plasma half-life extended from 3 minutes to 90 minutes. Treats acromegaly and neuroendocrine tumors. - **Enfuvirtide (Fuzeon)** — a 36-residue peptide that blocks HIV-1 entry by binding the gp41 fusion peptide. The first FDA-approved fusion inhibitor (2003). - **Ziconotide (Prialt)** — a synthetic ω-conotoxin. Blocks N-type calcium channels (Cav2.2) in spinal cord neurons. Delivered intrathecally for intractable pain. - **Tirzepatide (Mounjaro)** — a dual GIP/GLP-1 receptor agonist. A single 39-residue peptide activates two incretin receptors simultaneously. Achieved HbA1c reductions and weight loss exceeding those of semaglutide in head-to-head trials (SURMOUNT and SURPASS programs). --- ## Frontiers of Innovation ### Peptide–Drug Conjugates (PDCs) PDCs attach cytotoxic payloads to tumor-homing peptides. The peptide finds the target; the linker releases the drug inside the cell. Melflufen (melphalan flufenamide) used a dipeptide to deliver alkylating chemotherapy selectively to myeloma cells expressing aminopeptidases. The principle mirrors antibody–drug conjugates but with smaller carriers, faster tissue penetration, and lower immunogenicity. ### Stapled Peptides α-Helical peptides lose structure in solution and degrade within minutes. Hydrocarbon stapling — introduced by Gregory Verdine at Harvard — locks the helix by cross-linking non-adjacent residues with an olefin metathesis bridge. Stapled peptides resist proteolysis, cross cell membranes, and bind intracellular protein–protein interaction surfaces that small molecules cannot reach. ALRN-6924, a stapled peptide that disrupts the MDM2/MDMX–p53 interaction, entered clinical trials for wild-type p53 tumors. It reactivates the tumor suppressor pathway by preventing p53 degradation. ### Macrocyclic Peptides Cyclization constrains a peptide's conformation, pre-organizing it for target binding and shielding the amide backbone from proteases. The Random nonstandard Peptides Integrated Discovery (RaPID) system — developed by Hiroaki Suga at the University of Tokyo — combines mRNA display with genetic code reprogramming to screen libraries exceeding 10¹² unique macrocycles. RaPID has yielded sub-nanomolar inhibitors against targets previously considered undruggable, including protein–protein interfaces and flat binding surfaces. ### Cell-Penetrating Peptides (CPPs) Delivering cargo across lipid bilayers remains one of pharmacology's hardest problems. CPPs solve it. TAT peptide (from HIV-1 Tat protein, residues 47–57) and penetratin (from the Drosophila Antennapedia homeodomain) cross membranes by direct translocation or endocytosis and carry proteins, nucleic acids, and nanoparticles with them. Recent work conjugates CPPs to antisense oligonucleotides and siRNAs to silence disease genes without viral vectors. Sarepta Therapeutics' phosphorodiamidate morpholino oligomers (PMOs) conjugated to CPPs showed enhanced exon skipping in Duchenne muscular dystrophy preclinical models. ### Peptide-Based Vaccines Peptide epitopes train the immune system with precision. Unlike whole-pathogen vaccines, peptide vaccines present only the antigenic fragments that B cells and T cells recognize. This eliminates the risk of reversion or unintended immune activation. The approach gained momentum during COVID-19. EpiVacCorona (developed by Russia's Vector Institute) used synthetic peptides from the SARS-CoV-2 spike protein conjugated to a carrier. More broadly, neoantigen peptide vaccines — personalized cocktails of tumor-specific mutant epitopes — entered Phase II trials for melanoma, glioblastoma, and non-small-cell lung cancer. BioNTech's individualized neoantigen-specific immunotherapy (iNeST) platform, BNT122 (autogene cevumeran), showed promising results in pancreatic cancer when combined with atezolizumab and mFOLFIRINOX. ### AI-Driven Peptide Design Machine learning now designs peptides that evolution never explored. Key developments: - **RFdiffusion** (David Baker's lab, University of Washington) — generates novel protein and peptide backbones by denoising random coordinates through a diffusion model trained on known structures. Produces binders for specified targets with atomic-level accuracy. - **AlphaFold 2** (DeepMind) — predicts peptide–receptor complex structures, enabling rational design of peptide therapeutics against targets with known 3D conformations. - **ProteinMPNN** — designs amino acid sequences that fold into specified backbone geometries. Paired with RFdiffusion, it closes the loop from target structure to synthesizable peptide sequence. - **Generative language models** — large language models trained on protein sequences (ESM-2, ProGen) generate functional peptide sequences conditioned on desired properties: antimicrobial activity, membrane permeability, or target binding affinity. These tools compress discovery timelines from years to weeks. A 2023 study in *Nature Biotechnology* used RFdiffusion to design de novo peptide binders to influenza hemagglutinin that neutralized virus in cell-based assays — without any prior knowledge of existing antibody or peptide binders to that epitope. ### Peptide–Radiopharmaceutical Conjugates Peptides deliver radioactive isotopes directly to tumors for imaging and therapy. Lutetium-177 DOTATATE (Lutathera) — a somatostatin analogue chelated to ¹⁷⁷Lu — treats gastroenteropancreatic neuroendocrine tumors. The NETTER-1 trial demonstrated a 79% reduction in risk of disease progression. PSMA-targeting peptides labeled with ¹⁷⁷Lu (Pluvicto/lutetium-177 vipivotide tetraxetan) treat metastatic castration-resistant prostate cancer. The VISION trial showed improved overall survival. This class — peptide receptor radionuclide therapy (PRRT) — turns peptides into guided missiles that irradiate tumors from within. ### Self-Assembling Peptide Nanostructures Short peptides (as few as 2–3 residues) self-assemble into hydrogels, nanofibers, and nanotubes through non-covalent interactions — hydrogen bonding, π-stacking, and hydrophobic collapse. These materials serve as scaffolds for tissue engineering, depots for sustained drug release, and matrices for 3D cell culture. Diphenylalanine (FF) peptides assemble into rigid nanotubes with piezoelectric properties. RADA16-I forms hydrogels that support neural stem cell growth. The commercial product PuraStat (TDM-621), a self-assembling peptide hydrogel, achieves hemostasis during endoscopic procedures and received CE marking for clinical use. --- ## The Numbers The global peptide therapeutics market exceeded $45 billion in 2023 and grows at roughly 10% annually. Over 80 peptide drugs hold regulatory approval worldwide. More than 150 peptides occupy active clinical trials at any given time. GLP-1 receptor agonists alone — semaglutide, tirzepatide, liraglutide — generated over $30 billion in combined revenue in 2023, driven by the obesity and diabetes markets. --- ## Challenges That Remain Oral bioavailability limits most peptide drugs to injection. Gastric acid and pancreatic proteases destroy unprotected peptides within minutes. Rybelsus (oral semaglutide) overcame this barrier by co-formulating with SNAC (sodium N-[8-(2-hydroxybenzoyl)amino]caprylate), an absorption enhancer that raises local gastric pH and promotes transcellular transport. Bioavailability remains low (~1%), but the dose compensates. Designing peptides that survive the gut without permeation enhancers remains an open problem. Manufacturing cost scales with sequence length. Each coupling cycle in SPPS adds reagent expense and introduces cumulative losses. Flow chemistry and enzymatic ligation offer partial solutions. The field needs a step-change in synthesis efficiency to bring long peptides to cost parity with biologics produced by fermentation. Immunogenicity rises with repeated dosing. The immune system can generate anti-drug antibodies against synthetic peptides, particularly those with non-natural modifications. PEGylation, glycosylation, and sequence humanization mitigate but do not eliminate this risk. --- ## Summary Peptides occupy the pharmacological space between small molecules and large biologics. They bind targets with antibody-like specificity, penetrate tissues with small-molecule efficiency, and now — through stapling, cyclization, lipidation, and AI-guided design — overcome the metabolic instability that once confined them to niche applications. The convergence of computational design, synthetic chemistry, and clinical validation makes peptides one of the fastest-moving frontiers in drug development.