Peptide Science & Molecular Biology

N-terminal acetylation replaces the free amino group's positive charge with a neutral acetyl group (CH3CO-). In biology, approximately 80% of human proteins are N-terminally acetylated as a co-translational modification. For synthetic therapeutic peptides, N-acetylation serves two main purposes: protection from aminopeptidases that require a free N-terminal amino group, and modulation of receptor binding (some receptors bind better to acetylated peptides). In SPPS, N-acetylation is performed as a simple final step before cleavage from resin, using acetic anhydride or acetyl chloride.

Albumin has multiple drug-binding sites (Sudlow sites I and II) that accommodate diverse ligands through hydrophobic and electrostatic interactions. Fatty acid-modified peptides bind primarily to albumin's fatty acid binding sites. The binding is reversible — a dynamic equilibrium exists between bound (protected) and free (active) peptide. Only the free peptide can bind to its therapeutic receptor, creating a sustained-release depot effect. The fraction of drug bound to albumin at therapeutic concentrations (typically >99% for lipidated peptides) determines the effective free drug concentration and influences both efficacy and clearance. This pharmacokinetic property is thoroughly characterised during clinical development.

Approximately half of all bioactive peptide hormones in humans are C-terminally amidated. The amidation reaction is catalysed by the enzyme peptidylglycine alpha-amidating monooxygenase (PAM), which converts a C-terminal glycine-extended precursor to the amidated product. Amidation neutralises the negative charge of the C-terminal carboxyl group, which often improves receptor binding — many peptide receptors have evolved to recognise the amidated form. It also protects against carboxypeptidase degradation. In synthetic peptide manufacturing, C-terminal amidation is readily achievable using Rink amide resin during SPPS. Octreotide, goserelin, leuprolide, and many other approved peptides feature this modification.

Amino acids are linked together by peptide bonds to form chains. The sequence of amino acids determines the three-dimensional structure and function of the resulting peptide or protein. Each amino acid has an amino group (-NH2), a carboxyl group (-COOH), and a variable side chain (R group). The 20 standard amino acids are classified by their side chain properties: hydrophobic (e.g. leucine, valine), hydrophilic (e.g. serine, threonine), positively charged (e.g. lysine, arginine), and negatively charged (e.g. aspartate, glutamate). Non-standard amino acids, including D-amino acids and non-natural residues, are frequently incorporated into synthetic peptides to improve stability against enzymatic degradation.

Sequences are written from N-terminus to C-terminus using standard one-letter (e.g. HAEGTFTSDVSSYLEGQAAKEFIAWLVKGR for semaglutide's core) or three-letter codes (e.g. His-Ala-Glu-...). Even a single amino acid substitution can dramatically alter function — semaglutide differs from native GLP-1 at positions 8 (Aib substitution for DPP-4 resistance) and 34 (Arg to Arg with C-18 fatty diacid for albumin binding), extending half-life from 2 minutes to approximately one week. Sequence information is deposited in databases such as UniProt and is critical for understanding the relationship between natural peptides and their synthetic analogues.

Strategic substitutions serve multiple purposes in peptide drug design. Replacing L-amino acids with D-amino acids at protease-susceptible sites confers resistance to enzymatic cleavage. Substitution with alpha-aminoisobutyric acid (Aib, a non-natural amino acid) at position 8 of GLP-1 analogues prevents DPP-4 cleavage — this modification is used in semaglutide and other GLP-1 RAs. Conservative substitutions (replacing amino acids with chemically similar alternatives) can fine-tune receptor binding affinity and selectivity. Non-conservative substitutions may create entirely new pharmacological profiles. Structure-activity relationship (SAR) studies systematically evaluate how each position in the sequence contributes to biological activity.

Peptide analogues are designed by making strategic modifications to a natural peptide template. Common modification strategies include: amino acid substitutions to improve receptor binding or resist enzymatic cleavage; addition of chemical moieties (fatty acids, PEG chains) for half-life extension; truncation to identify the minimum active sequence; and cyclisation for conformational stability. Semaglutide is an analogue of GLP-1(7-37) with two key modifications: Aib at position 8 (DPP-4 resistance) and a C-18 fatty diacid via a linker at position 26 (albumin binding). Tesamorelin is a GHRH analogue with a trans-3-hexenoic acid modification. The development of improved analogues is a central strategy in peptide therapeutics.

The C-terminus (carboxyl terminus) is the last amino acid added during protein synthesis. The free carboxyl group is susceptible to carboxypeptidase degradation. C-terminal amidation — converting -COOH to -CONH2 — is one of the most common modifications in natural peptide hormones (found in approximately half of all bioactive peptides) and often increases receptor binding affinity and biological potency. Many therapeutic peptides (including octreotide and several GnRH compounds) feature C-terminal amidation. This modification also provides resistance to carboxypeptidase degradation.

Albumin is the most abundant plasma protein (approximately 35-50 g/L) with a half-life of approximately 19 days. Its abundance and long half-life make it an ideal carrier for peptide drugs designed with albumin-binding modifications. Beyond non-covalent albumin binding (semaglutide, liraglutide, somapacitan), some drug design strategies use direct covalent albumin conjugation or albumin-binding domains fused to therapeutic peptides. The albumin recycling pathway (via the neonatal Fc receptor, FcRn) contributes to the long circulation time of albumin-bound drugs. Romiplostim uses a different carrier strategy — it is fused to an Fc antibody fragment for extended half-life.

Cyclisation occurs through several mechanisms: head-to-tail (N-terminus to C-terminus backbone linkage), side chain-to-side chain (e.g. disulphide bonds between cysteines, lactam bridges between lysine and glutamate), or backbone-to-side chain connections. Cyclic peptides resist enzymatic degradation better than linear peptides because they lack free terminal ends recognised by exopeptidases, and their constrained conformation reduces endopeptidase access. Cyclosporine (11 amino acids) is a cyclic peptide essential to immunosuppressive therapy. Octreotide, lanreotide, and pasireotide are cyclic somatostatin analogues. The cyclic scaffold is actively exploited in modern peptide drug design programmes.

Natural proteins are composed exclusively of L-amino acids (left-handed stereochemistry). D-amino acids (right-handed) are their mirror images. Since human proteases have evolved to recognise and cleave L-amino acid sequences, D-amino acid incorporation at key positions creates peptide bonds that resist enzymatic degradation. D-amino acids can also alter the peptide's secondary structure and receptor binding properties. Certain D-amino acid substitutions are tolerated at non-critical binding positions while dramatically improving metabolic stability. Some naturally occurring antimicrobial peptides from bacteria (e.g. gramicidin) contain D-amino acids as a natural resistance mechanism against host proteases.

Disulphide bonds form through oxidation of thiol (-SH) groups on cysteine residues. They are important structural features in many peptide hormones: oxytocin has one intramolecular disulphide bond forming a 20-membered ring, vasopressin has a similar structure, and insulin has two interchain and one intrachain disulphide bond. The presence of disulphide bonds affects manufacturing (correct bond formation must be ensured during synthesis) and stability (reducing conditions in the GI tract can break these bonds, contributing to poor oral bioavailability). In quality control, disulphide bond integrity is verified through analytical methods including mass spectrometry and peptide mapping.

The human body produces hundreds of endogenous peptides serving as hormones (GLP-1, GHRH, GnRH, somatostatin), neurotransmitters (substance P, enkephalins, endorphins), paracrine signals (growth factors), and antimicrobial defence molecules (defensins, cathelicidins). Some therapeutic peptides are bio-identical to their endogenous counterparts — synthetic oxytocin and vasopressin have identical amino acid sequences to the natural hormones. Others are modified analogues designed to improve upon natural pharmacological properties. Understanding the endogenous peptide landscape is essential for evaluating the safety profile of therapeutic peptides, as bio-identical compounds may have more predictable effects.

Major endopeptidases include the serine proteases trypsin (cleaves after Arg, Lys), chymotrypsin (cleaves after Phe, Trp, Tyr), and elastase (cleaves after small residues). In the bloodstream, neprilysin (EC 3.4.24.11) is a membrane-bound metalloendopeptidase that degrades numerous bioactive peptides. Pepsin in the stomach degrades peptides at low pH. Identifying the endopeptidase-susceptible sites in a therapeutic peptide sequence enables targeted modifications to improve stability. Cyclisation eliminates terminal cleavage sites but does not protect against endopeptidase cleavage within the ring — additional modifications may be needed at vulnerable internal positions.

When exogenous peptides are administered, the body processes them through the same proteolytic and clearance mechanisms that handle endogenous peptides. Exogenous peptides that are structurally identical to endogenous counterparts (e.g. oxytocin, vasopressin, glucagon) generally integrate into existing physiological pathways. Modified analogues must be evaluated for novel interactions — their structural alterations may produce effects not seen with the natural molecule. The distinction between endogenous and exogenous is also relevant to anti-doping regulations, where synthetic versions of endogenous peptide hormones (such as growth hormone) are prohibited in competitive sports.

Aminopeptidases remove amino acids from the N-terminus (aminopeptidase N, leucine aminopeptidase) while carboxypeptidases remove them from the C-terminus (carboxypeptidase A, B, E). DPP-4 is technically an exopeptidase — it removes a dipeptide from the N-terminus. Terminal modifications (N-acetylation blocks aminopeptidases; C-amidation blocks carboxypeptidases; Aib at position 2 blocks DPP-4) are the most straightforward protective strategies. Cyclic peptides inherently lack susceptible termini. The relative contribution of exopeptidase vs endopeptidase degradation varies by peptide and determines which stabilisation strategy is most effective.

Glycosylation involves attachment of sugar chains (glycans) to asparagine residues (N-linked) or serine/threonine residues (O-linked). Glycans affect protein folding, stability, solubility, half-life, and immunogenicity. For recombinant therapeutic peptides produced in mammalian cell systems, glycosylation patterns depend on the host cell type and culture conditions, which is why glycosylation consistency is a major quality concern for biological products. Glycosylation differences can affect biosimilar comparability assessments. Some naturally glycosylated hormones (FSH, LH, TSH, hCG) require mammalian cell production to achieve appropriate glycosylation patterns.

Multiple isoforms arise from alternative mRNA splicing, post-translational modifications, or allelic variation. Growth hormone has a predominant 22 kDa isoform (approximately 90% of circulating GH) and a 20 kDa isoform (approximately 10%) from alternative splicing of the same gene. These isoforms have different receptor binding properties and biological activities. Recombinant somatropin is the 22 kDa isoform. IGF-1 exists in multiple isoforms due to alternative splicing and differential glycosylation. Understanding isoform biology is important because drug development targets specific isoforms, and analytical methods must distinguish between them during quality control.

Lipidation exploits the reversible binding of fatty acid-modified peptides to albumin in the bloodstream. When the peptide-albumin complex circulates, the large size of albumin (approximately 67 kDa) prevents renal filtration, and the bound state shields the peptide from proteolytic enzymes. Semaglutide uses a C-18 fatty diacid attached via a mini-PEG linker at Lys26 to achieve albumin binding, extending its half-life to approximately 165 hours (enabling weekly dosing). Liraglutide uses a C-16 fatty acid (palmitic acid) at Lys26, achieving a half-life of approximately 13 hours (daily dosing). This albumin-binding lipidation platform has become a key technology in long-acting peptide drug development.

Molecular weight is calculated by summing atomic weights of all constituent atoms and is measured in daltons (Da) or kilodaltons (kDa). Small peptides range from a few hundred Da (e.g. TRH at 362 Da) to several thousand. Larger therapeutic peptides approach or exceed protein range: somatropin is approximately 22,124 Da, insulin approximately 5,808 Da, and semaglutide approximately 4,114 Da. Molecular weight affects pharmacokinetic properties including renal filtration (molecules above approximately 60,000 Da are not efficiently filtered by the kidneys), cellular permeability, and analytical method selection. It also determines whether chemical synthesis or recombinant production is more appropriate.

The N-terminus (amino terminus) is where peptide synthesis begins in biological systems (ribosomes read mRNA 5' to 3', producing proteins from N to C). In solid-phase peptide synthesis, however, construction typically proceeds from C-terminus to N-terminus. The free amino group at the N-terminus makes it susceptible to aminopeptidases — enzymes that sequentially remove amino acids from this end. N-terminal modifications such as acetylation or pyroglutamate formation are common strategies to protect therapeutic peptides from aminopeptidase degradation. DPP-4, which degrades GLP-1, cleaves at the penultimate position from the N-terminus.

Neuropeptides are produced and released by neurons and typically act on G-protein coupled receptors. They differ from classical neurotransmitters (dopamine, serotonin, glutamate) in several ways: they are larger molecules (3-40+ amino acids vs single amino acids or small molecules), synthesised in the cell body and transported to nerve terminals (rather than synthesised locally), stored in large dense-core vesicles, and typically produce slower-onset but longer-lasting effects. Over 100 neuropeptides have been identified. Relevant examples include oxytocin, vasopressin, substance P (pain), neuropeptide Y (appetite), endorphins (pain/reward), and CGRP (migraine). Several research compounds target neuropeptide systems for cognitive or neuroprotective applications.

PEG chains range from 5 to 40+ kDa and increase the molecule's hydrodynamic radius, reducing glomerular filtration by the kidneys and creating a hydrophilic shield against proteolytic enzymes. The degree and site of PEGylation must be carefully optimised — over-PEGylation can sterically hinder receptor binding and reduce biological activity. Palopegteriparatide is a PEGylated PTH that provides sustained parathyroid hormone activity. Somapacitan uses a different half-life extension approach (albumin binding via a C-18 fatty acid) rather than PEGylation. Concerns about PEG immunogenicity (anti-PEG antibodies) and PEG accumulation in tissues have led to exploration of alternative polymer conjugation strategies.

Peptidases are functionally classified as endopeptidases (cleaving internal bonds) and exopeptidases (cleaving terminal residues). Exopeptidases include aminopeptidases (N-terminal cleavage), carboxypeptidases (C-terminal cleavage), and dipeptidyl peptidases (removing dipeptides from the terminus). DPP-4 is a dipeptidyl aminopeptidase — it removes a dipeptide from the N-terminus of substrates with alanine or proline at position 2. Neprilysin (neutral endopeptidase) degrades several endogenous peptides including natriuretic peptides, bradykinin, and substance P. The collective activity of peptidases determines the metabolic fate and half-life of circulating peptide drugs.

Peptides are classified by length: dipeptides (2 amino acids), tripeptides (3), oligopeptides (fewer than 20), and polypeptides (20-50, though the boundary with proteins is not precise). Many naturally occurring hormones are peptides, including insulin (51 aa), oxytocin (9 aa), and growth hormone-releasing hormone (44 aa). Therapeutic peptides may be bio-identical to natural peptides (e.g. synthetic oxytocin), modified analogues designed for improved pharmacological properties (e.g. semaglutide), or entirely synthetic sequences designed to interact with specific biological targets. The peptide therapeutics market has grown substantially, driven by advances in synthesis technology and delivery systems.

Peptide bonds have partial double-bond character due to resonance, which gives the bond a planar, rigid structure. This rigidity constrains the three-dimensional folding of peptide chains. The bond length is approximately 1.33 Å, intermediate between a single C-N bond (1.49 Å) and a double C=N bond (1.27 Å). Peptide bonds are cleaved by proteolytic enzymes (proteases) in the body, which is why many therapeutic peptides have short half-lives and require injection. Modifications such as N-methylation, incorporation of D-amino acids, cyclisation, or use of reduced peptide bonds (CH2-NH) can protect peptide bonds from enzymatic degradation.

Correct folding is driven by the amino acid sequence (Anfinsen's dogma — the sequence contains all information needed for folding) and influenced by environmental conditions (pH, temperature, ionic strength, redox state). Misfolded peptides may be biologically inactive, prone to aggregation, or immunogenic. In recombinant production, misfolding can lead to inclusion body formation in bacterial hosts, requiring denaturation and refolding steps. For synthetic peptides with disulphide bonds, oxidative folding must produce the correct disulphide bond pattern among multiple possible arrangements. Quality control methods including circular dichroism and NMR spectroscopy can assess peptide folding.

Peptide libraries are generated through combinatorial chemistry (solid-phase synthesis with split-and-mix approaches), phage display (expressing peptides on bacteriophage surfaces), mRNA display, ribosome display, or computational design. Libraries can contain 10^6 to 10^13 unique sequences. Screening methods include affinity-based selection (biopanning against immobilised targets), high-throughput functional assays, and in silico docking simulations. Hits identified from library screens serve as starting points for medicinal chemistry optimisation — lead peptides are iteratively modified to improve potency, selectivity, stability, and pharmacokinetics. This process has been used to discover many clinically relevant peptide drug candidates.

Sequence homology is determined by aligning two peptide sequences and calculating the percentage of identical (or similar) amino acid positions. High homology to an endogenous peptide suggests the therapeutic analogue will interact with the same receptors and pathways. Lower homology indicates more extensive modification. Exenatide (from Gila monster exendin-4) shares approximately 53% sequence homology with human GLP-1, while liraglutide (a modified human GLP-1) shares approximately 97% homology. Homology also affects immunogenicity risk — peptides with lower homology to human sequences are more likely to be recognised as foreign by the immune system.

Peptidomimetics retain key structural features (pharmacophores) enabling receptor binding but replace the peptide backbone with more stable chemical scaffolds. Strategies include: backbone modification (N-methylation, reduced amide bonds, peptoids with side chains on nitrogen), non-natural amino acid incorporation, beta-peptides (using beta-amino acids with an extra methylene group), and entirely non-peptide small molecules designed from peptide pharmacophore templates. Some approved drugs classified as peptides have peptidomimetic features. Relugolix and elagolix are orally bioavailable GnRH antagonists that represent the extreme end of peptidomimetic design — they are small molecules derived from peptide receptor pharmacology.

In phage display, synthetic DNA encoding random peptide sequences is inserted into the gene for a bacteriophage coat protein (typically pIII or pVIII of M13 phage). Each phage particle displays a unique peptide on its surface while carrying the encoding DNA internally. The library (typically 10^8-10^10 variants) is screened by exposing phages to an immobilised target — those that bind are retained, amplified, and reselected over multiple rounds (biopanning). DNA sequencing identifies the winning peptide sequences. George Smith and Sir Gregory Winter shared the 2018 Nobel Prize in Chemistry for phage display technology. It has been used to discover peptide drug leads and antibody therapeutics.

The distinction between peptide and polypeptide is based on chain length, though the boundary is not universally agreed. Chains of 2-20 amino acids are generally called peptides, while longer chains are polypeptides. When a polypeptide folds into a functional three-dimensional structure (often with multiple domains), it is typically referred to as a protein. Many therapeutic compounds tracked on PeptideTrace — such as somatropin (191 amino acids) and glatiramer acetate (a mixture of polypeptides averaging 40-100 residues) — are technically polypeptides or proteins but are commonly discussed alongside shorter peptides due to shared biological mechanisms.

Common PTMs include: phosphorylation (addition of phosphate groups, key in cell signalling), glycosylation (attachment of sugar chains, affects folding and stability), acetylation (addition of acetyl groups), methylation, hydroxylation (critical for collagen stability — requires vitamin C), amidation (C-terminal modification common in bioactive peptides), and disulphide bond formation. In therapeutic peptide development, PTMs can be deliberately introduced or prevented to optimise drug properties. PEGylation and lipidation are artificial PTMs applied to therapeutic peptides to extend half-life. Understanding which PTMs are present on endogenous peptides helps explain why some synthetic analogues differ from their natural templates.

Proteases (also called peptidases or proteinases) are classified by their catalytic mechanism: serine proteases (e.g. trypsin, chymotrypsin, DPP-4), cysteine proteases (e.g. caspases), aspartic proteases (e.g. pepsin, renin), metalloproteases (e.g. matrix metalloproteinases, angiotensin-converting enzyme), and threonine proteases (the proteasome's catalytic subunits). DPP-4 is a serine protease of particular importance in peptide therapeutics because it rapidly degrades GLP-1 and GIP. Understanding which proteases degrade a specific therapeutic peptide enables rational design of resistant analogues. Protease inhibitors (e.g. bortezomib, carfilzomib) are themselves therapeutic agents targeting the proteasome.

Proteins are distinguished from peptides by size and structural complexity. While peptides are generally short chains under 50 amino acids, proteins are larger molecules that adopt stable three-dimensional conformations through secondary structure (alpha-helices, beta-sheets), tertiary structure (overall 3D shape), and sometimes quaternary structure (multi-subunit assemblies). Some therapeutic compounds, such as somatropin (22,124 Da), are proteins rather than peptides. This distinction matters for manufacturing — proteins typically require recombinant DNA technology rather than chemical synthesis — and for regulatory classification under the BLA rather than NDA pathway.

Proteolysis occurs at multiple sites in the body: the gastrointestinal tract (pepsin, trypsin, chymotrypsin), blood plasma (various circulating proteases), liver, kidneys, and within cells (lysosomal proteases, proteasome). For peptide drugs, proteolysis determines half-life and dosing frequency. Strategies to resist proteolysis include: D-amino acid incorporation, N-terminal acetylation, C-terminal amidation, backbone modifications (N-methylation, reduced peptide bonds), cyclisation, PEGylation, lipidation for albumin binding, and formulation in sustained-release systems. The specific protease(s) responsible for degrading each therapeutic peptide are typically identified during preclinical development to guide stabilisation strategies.

Recombinant production involves inserting the gene encoding the peptide into a host organism (commonly E. coli, Saccharomyces cerevisiae yeast, or Chinese hamster ovary mammalian cells) that reads the genetic instructions and produces the peptide using its own cellular machinery. This method is used for peptides too large or structurally complex for chemical synthesis, typically exceeding 40-50 amino acids. Somatropin is the classic example — identical to 191 amino acid natural human growth hormone but produced in bacterial or mammalian cells. Recombinant peptides require extensive purification to remove host cell proteins, DNA, and endotoxins. Manufacturing is subject to strict cGMP standards and bioprocess validation.

Solid-phase peptide synthesis (SPPS) builds peptide chains one amino acid at a time on a solid resin support using Fmoc chemistry as the standard approach. This method allows precise sequence control and incorporation of non-natural amino acids, D-amino acids, and chemical modifications impossible with recombinant production. Most approved peptide drugs under 50 amino acids are chemically synthesised: semaglutide (31 aa + modifications), octreotide (8 aa), leuprolide (9 aa), goserelin (10 aa), and teriparatide (34 aa). Research compounds are almost universally produced by chemical synthesis. Purity is assessed by HPLC and identity confirmed by mass spectrometry.

Tertiary structure is determined by interactions between amino acid side chains: hydrophobic interactions (non-polar residues cluster in the molecule's interior), hydrogen bonds, ionic bonds (salt bridges between charged residues), van der Waals forces, and disulphide bonds between cysteine residues. For therapeutic peptides, tertiary structure directly affects receptor binding — the peptide must adopt the correct conformation to fit into the receptor binding site. Longer peptides and proteins have more complex tertiary structures. Manufacturing processes must preserve correct tertiary structure; denaturation (unfolding) can result in loss of biological activity and increased immunogenicity.