What Are Peptides? Structure, Synthesis and Classes

Peptides and proteins are built from the same twenty amino acids, joined by the same chemical bond, and separated only by an arbitrary line drawn somewhere around fifty residues. Nothing chemical happens when a molecule crosses that line, which is exactly why the word peptides causes so much confusion in catalogs, literature searches and procurement documents.

What are peptides?

Peptides are short chains of amino acids linked by peptide bonds, conventionally containing between two and roughly fifty residues. Longer chains are called proteins. Peptides function as hormones, signaling molecules and antimicrobial agents, and research-grade peptides are produced mainly by solid phase synthesis, then verified by reversed-phase HPLC and mass spectrometry.

All compounds discussed on this page are sold for laboratory research use only and are not for human consumption.

The most common oversimplification is treating peptides as a product category rather than a chemical class. A peptide is defined by its bond chemistry and chain length, not by the research area a listing associates it with, which is why several widely stocked research compounds are not peptides at all. This page covers peptide bond chemistry and the size convention, how peptides are synthesised, the functional classes described in the peer-reviewed literature, the analytical methods used to confirm identity, and what research use only status means for laboratory procurement.

What exactly is a peptide at the molecular level?

A peptide is a polymer of amino acid residues joined by amide linkages called peptide bonds. Each of the twenty proteinogenic amino acids shares the same backbone architecture: a central alpha carbon carrying an amino group, a carboxyl group, a hydrogen, and a variable side chain that gives the residue its chemistry.

The peptide bond forms by condensation. The alpha-carboxyl group of one amino acid reacts with the alpha-amino group of the next, releasing water and leaving a C-N amide linkage. That is why a peptide's theoretical mass is not simply the sum of its amino acid masses: one water molecule is lost per bond formed.

The resulting chain has direction. One end carries a free amino group, the N-terminus, and the other a free carboxyl group, the C-terminus, and sequences are always written N to C. That convention matters because a sequence read backwards describes a completely different molecule.

The amide bond has partial double bond character, so the backbone cannot rotate freely around the C-N linkage. The unit stays planar and overwhelmingly in the trans configuration, which constrains the chain into the phi and psi backbone angles that generate helices, turns and extended strands. Short peptides are often flexible and disordered in solution and only adopt a defined shape when they engage a binding partner. A 2018 review of the peptide chemistry toolbox describes how that flexibility is both the functional advantage and the stability weakness of the class (PubMed PMID 29395804).

Where is the line between a peptide and a protein?

The boundary is a convention, not a physical property. Chains of roughly two to twenty residues are usually called oligopeptides, chains up to about fifty residues are peptides, and anything longer is generally called a protein or a polypeptide, with a rough molecular weight cutoff near five kilodaltons.

Insulin shows how soft the line is. At 51 residues across two disulfide-linked chains, it is described as a peptide hormone in endocrinology texts and as a small protein in structural biology papers, and both are correct.

The practical consequences are real even if the definition is not. Chain length determines how a molecule is made: standard solid phase synthesis is routine for short sequences and gets progressively harder as the chain extends, which pushes longer targets toward chemical ligation strategies or recombinant expression. Chain length also shapes analysis, since a short sequence can be sequenced directly by tandem mass spectrometry, while larger molecules often require enzymatic digestion first. A 2025 review of classical and emerging synthesis methods sets out how method selection tracks with target size and complexity (PubMed PMID 41191975).

How do sequence and chemical modification change what peptides do?

Primary sequence sets function, but terminal and side chain chemistry sets survivability. Native peptides are cleared quickly in biological matrices because exopeptidases attack free termini and endopeptidases cut at predictable internal motifs, with half-lives measured in minutes rather than hours.

The medicinal chemistry responses are well catalogued. N-terminal acetylation and C-terminal amidation remove charged termini that exopeptidases recognise. Disulfide bridges between cysteine residues lock a sequence into a constrained ring, as in somatostatin and oxytocin. Head-to-tail cyclisation and lactam bridges rigidify the backbone. Substituting a D-amino acid or a non-natural residue at a cleavage site blocks proteases that only recognise L-configuration substrates. Lipidation attaches a fatty acid chain that drives reversible albumin binding, and polyethylene glycol conjugation increases hydrodynamic radius and slows renal filtration.

These are not cosmetic edits. A single modification can change solubility, aggregation behaviour, receptor selectivity and storage stability simultaneously. Reviews of peptide-based drug discovery trace how stabilisation chemistry, rather than sequence discovery, has driven most of the field's progress (PubMed PMID 36481586, PubMed PMID 28720325).

For a laboratory, the implication is narrower: two vials labelled with the same sequence name can contain chemically distinct molecules if one is amidated and the other is not. The molecular formula and theoretical mass on the certificate, not the common name, settle which one is present.

How are research peptides actually synthesised?

Most research peptides are made by solid phase peptide synthesis, the method Robert Bruce Merrifield introduced in 1963 and for which he received the Nobel Prize in Chemistry in 1984. The growing chain stays anchored to an insoluble resin bead, which means excess reagents and byproducts can be washed away at every stage.

The chain is assembled from the C-terminus toward the N-terminus. The first residue is attached to the resin, its temporary alpha-amino protecting group is removed, the next protected amino acid is activated and coupled, and the sequence repeats residue by residue. Two chemistries dominate: Fmoc/tBu, where the fluorenylmethyloxycarbonyl group is removed with piperidine under mild basic conditions, and the older Boc/Bzl approach, which uses acid for each deprotection step. Fmoc chemistry has become the default for most research-scale work (PubMed PMID 26785684).

Coupling relies on activating reagents such as carbodiimides with Oxyma, or uronium salts including HBTU and HATU, which convert the carboxyl group into a reactive ester. After the final residue, concentrated trifluoroacetic acid cleaves the peptide from the resin and strips the side chain protecting groups, after which the crude material is precipitated, purified by preparative reversed-phase chromatography, and lyophilised.

Failure modes include on-resin aggregation in hydrophobic stretches, aspartimide formation at Asp-Gly motifs, racemisation at activated residues, and truncated sequences that differ from the target by a single residue. Each produces a related impurity that can be chemically similar enough to co-elute without careful chromatographic method development.

Two other routes matter. Recombinant expression handles longer chains and complex folds more economically than stepwise synthesis. And there is active work on reducing the solvent burden of solid phase synthesis, reviewed in a 2019 survey of greener synthesis approaches (PubMed PMID 30681290).

Which functional classes do peptides fall into?

The literature sorts peptides by what they do, and five groupings cover most of what laboratories work with.

Peptide hormones and endocrine regulators. Insulin, glucagon, somatostatin and the natriuretic peptide family sit here. Somatostatin signals through G protein-coupled receptors whose structure and function are mapped in a 2025 Endocrine Reviews article (PubMed PMID 39116368), and the natriuretic peptides ANP, BNP and CNP act through guanylyl cyclase-linked receptors described in a 2009 handbook review (PubMed PMID 19089336). The nonapeptide oxytocin belongs to the same broad grouping, with a disulfide-constrained ring and an amidated C-terminus.

Releasing factors and secretagogue analogs. Growth hormone-releasing hormone analogs such as tesamorelin and CJC-1295 are modified fragments of the native 44-residue sequence, while ipamorelin is a pentapeptide studied as a ghrelin receptor ligand. MK-677 is grouped with them in catalogs but is a non-peptide small molecule.

Antimicrobial peptides. These are typically cationic and amphipathic, folding so that charged and hydrophobic faces separate, which allows them to interact with anionic microbial membranes. The human cathelicidin LL-37 and the defensins are the standard references. Structure-function work on this class has concentrated on net charge, helicity and hydrophobic moment rather than receptor binding (PubMed PMID 9923692, PubMed PMID 27872334).

Immune and inflammation-related signaling peptides. Thymosin alpha-1 is a 28-residue acetylated peptide originally isolated from thymic tissue. KPV is a tripeptide corresponding to the C-terminal sequence of alpha-melanocyte-stimulating hormone.

Fragment and repair-associated research peptides. BPC-157 is a pentadecapeptide sequence derived from a gastric protein, TB-500 corresponds to an active region of thymosin beta-4, GHK-Cu is the copper-binding tripeptide glycyl-L-histidyl-L-lysine, and AOD-9604 is a modified fragment of the C-terminal region of human growth hormone. MOTS-c belongs to the newer mitochondrial-derived peptide class, encoded within mitochondrial DNA. Epithalon is a synthetic tetrapeptide studied in gerontology literature.

Neuropeptides and larger synthetic analogs. Semax and Selank are short synthetic analogs of endogenous regulatory sequences studied in neuroscience research, a field whose conceptual framework for stress signaling was revisited in a 2020 Journal of Neuroscience article (PubMed PMID 31896560). At the long end of the size range sit the incretin receptor agonists studied in metabolic research programs, including semaglutide, tirzepatide, retatrutide and survodutide, which combine 30-plus residue chains with lipid side chains for stability.

Here is the part most catalog pages skip. Several compounds routinely grouped with peptides are chemically unrelated to them. NAD+ is a dinucleotide coenzyme built from nicotinamide and adenine ribonucleotides, with no amide backbone of amino acids. 5-Amino-1MQ is a small-molecule quinolinium compound. SLU-PP-332 is a synthetic small-molecule agonist. None contain a peptide bond.

How do the major peptide classes compare?

The table below compares the main functional groupings by chain length, defining chemistry and the depth of published literature behind each, which is the variable that most often decides whether a compound is practical to work with.

How do laboratories confirm a peptide is what the label says?

Identity and composition are established by orthogonal methods, because no single technique answers both questions. Reversed-phase HPLC on a C18 stationary phase with a water and acetonitrile gradient and an acidic modifier separates the target from related impurities and gives a retention time and a peak area profile. Detection is usually at 214 nanometres, where the peptide bond absorbs, and often simultaneously at 280 nanometres, which responds to tryptophan and tyrosine.

Mass spectrometry answers the identity question. Electrospray ionisation or MALDI-TOF measurement is compared against the theoretical monoisotopic and average masses calculated from the sequence, and tandem MS fragmentation can confirm the actual residue order. A mass that is 18 daltons off points toward an unintended water loss or gain; a mass 1 dalton low on a sequence expected to be amidated points to a free acid C-terminus instead.

This is where the single most common misreading happens. Chromatographic peak area and peptide content are not the same measurement, and a lyophilised vial can produce a visually clean chromatogram while a substantial share of the physical mass in the vial is water, counterion salt and residual solvent. Trifluoroacetic acid from the cleavage and purification stages typically remains as the counterion unless it has been exchanged for acetate, and basic residues such as arginine and lysine each bind counterion mass. Net peptide content is established by amino acid analysis after acid hydrolysis or by nitrogen determination, water by Karl Fischer titrimetry, and counterion by ion chromatography. A certificate that reports only a chromatogram has answered one question and left the other open.

Supporting checks include residual solvent analysis, bacterial endotoxin testing where the application requires it, appearance of the lyophilised cake, and confirmation of the CAS number and molecular formula against the sequence before procurement. Real Peptides publishes lot-specific certificates of analysis for the compounds it supplies, and a separate reference page explains how to read a peptide certificate of analysis line by line.

What does research use only mean for peptides?

Research use only is a regulatory status, not a quality grade. It means the material is supplied and labelled for laboratory investigation and is not an approved drug product for human or veterinary use, has not been evaluated by any medicines regulator for safety or efficacy in people, and carries no approved labelling, indication or usage instructions.

That status has concrete operational consequences. A research-use-only supplier documents what is in the vial and does not provide usage, preparation or handling guidance aimed at a person, because any such guidance would reframe a laboratory chemical as something it is not. Real Peptides supplies research-use-only compounds and publishes lot certificates of analysis for them, and that is the limit of what a legitimate supplier in this category asserts.

It also means that the burden of characterisation shifts to the purchaser. There is no regulatory batch release system standing behind research-grade peptides the way there is behind an approved medicine, so the certificate of analysis, the analytical methods behind it, and the traceability of the lot are the documentation a laboratory actually has. Peptides are reactive, hygroscopic and sensitive to oxidation at methionine, cysteine and tryptophan residues, which is why cold-chain handling and documented storage conditions are part of material quality.

What If: Common Peptide Verification Scenarios

What if the observed mass differs from the theoretical molecular weight?

Treat any mismatch as an identity question until the difference is accounted for arithmetically. A deviation of 18 daltons usually indicates an unintended hydration or dehydration event, 1 dalton can separate an amidated C-terminus from a free acid, and 2 daltons commonly reflects disulfide bond formation or reduction. Differences matching a single residue mass point toward a deletion sequence from an incomplete coupling step during synthesis.

What if the chromatogram shows a shoulder on the main peak?

A shoulder means co-eluting material that the current separation method cannot resolve, not necessarily a failed synthesis. Related impurities such as deletion sequences, aspartimide-derived byproducts and epimers are chemically close to the target and often require a modified gradient, a different column chemistry or an altered mobile phase modifier to separate. The relevant impurity families are catalogued in the reference page on peptide impurities.

What if a peptide will not go into solution in an aqueous buffer?

Solubility behaviour is predictable from the sequence rather than from the compound name. Peptides are least soluble near their isoelectric point, where net charge approaches zero, and sequences dominated by hydrophobic or beta-sheet-forming residues aggregate readily. Calculating net charge from the count of acidic and basic residues at the working pH usually explains the behaviour.

What if a compound listed with peptides turns out not to contain a peptide bond?

Check the molecular formula and CAS number against the stated structure before the material enters an inventory or a literature search. Catalog grouping follows research interest, not chemistry, which is how dinucleotides and small-molecule agonists end up listed next to genuine peptides. Misclassification propagates into search strategies, storage decisions and analytical method selection.

What does the peptide literature actually support right now?

Here is the honest answer: the depth of evidence across peptide classes is wildly uneven, and treating all peptides as a single evidentiary category is the central error in this field. Endocrine peptides have decades of receptor-level structural and physiological work behind them. Antimicrobial peptides have well-mapped biophysical mechanisms. A large share of the fragment-derived compounds that dominate research catalogs rest on comparatively thin preclinical literature, with mechanisms described in a handful of papers rather than established across independent laboratories. That gap is not a reason to ignore those compounds in research settings. It is a reason to describe them accurately, cite what exists, and avoid implying more than the published record supports.

How was this page researched?

This page was compiled from peer-reviewed literature indexed in PubMed and from standard peptide chemistry references. No first-party laboratory testing, internal data or unpublished results informed any statement here.

Structural and synthesis content draws on reviews of Fmoc solid phase synthesis (PubMed PMID 26785684), classical and emerging synthesis methods (PubMed PMID 41191975), greener synthesis approaches (PubMed PMID 30681290) and the peptide chemistry toolbox (PubMed PMID 29395804). Class descriptions draw on reviews of somatostatin receptor biology (PubMed PMID 39116368), natriuretic peptides (PubMed PMID 19089336), antimicrobial peptide mechanisms and structure-function relationships (PubMed PMID 27872334, PubMed PMID 9923692) and the stress concept in neuroscience (PubMed PMID 31896560). Field-level context comes from reviews of peptide-based drug discovery (PubMed PMID 36481586), historical development of therapeutic peptides (PubMed PMID 28720325), peptide-based delivery systems and delivery system overviews (PubMed PMID 34833427, PubMed PMID 31435914, PubMed PMID 31848464), a 2025 review of peptide development platforms (PubMed PMID 40038239) and a 2026 clinical primer summarising peptide literature for physicians (PubMed PMID 41476424). Each source is described only as far as its title and scope support.

Related reference pages on this site cover what research use only means, what defines pure peptides, the types of impurities found in synthetic peptides, and company and facility information.

The most useful thing to carry away about peptides is that the name on a vial is a label, not a measurement. Sequence, molecular formula, theoretical mass and counterion identity are the only properties that survive contact with an analytical instrument, and every meaningful question about a peptide, whether it is the right molecule, whether it will stay intact in storage, whether it will dissolve, resolves back to those four facts. Chemistry does not care what a compound is popularly associated with. It cares what bonds are present and in what order.