What is a peptide bond? Peptide structure, chemistry and synthetic peptides explained

What is the definition of a peptide in biology and in medicine?

In biology, a peptide is a chain of amino acids joined by peptide bonds. In medicine the word names the same chain when it acts as a hormone, a signal or a medicine. The Therapeutic Goods Administration (TGA) defines peptides as short chains of amino acids, the building blocks of proteins, many of which exist naturally in the body and take part in signalling (TGA peptide products page, 23 September 2026). The National Library of Medicine's MeSH vocabulary indexes peptides as a general term and sets oligopeptides at fewer than 13 amino acids, while peptide structure databases use fewer than 50 (MeSH record for Peptides, checked 14 October 2026; Hamdani et al., Protein Science, 2026).

The medical meaning leans on function. Wang and colleagues describe therapeutic peptides as well-ordered amino acids, usually with molecular weights of 500 to 5,000 daltons, and trace the field to natural hormones such as insulin, oxytocin, vasopressin and gonadotropin-releasing hormone (Wang et al., Signal Transduction and Targeted Therapy, 2022). More than 80 peptide drugs have been approved worldwide (Wang et al., 2022). The page on what peptides are, and which are medicines, has the full answer.

No source sets one universal line between a peptide and a protein. Hamdani and colleagues point out that the fewer-than-50 rule used by databases mixes together very different cases (Hamdani et al., 2026). The table below sets out the size conventions each source uses.

What is a peptide bond in simple terms, and where are peptide bonds?

A peptide bond is the link between two neighbouring amino acids in a chain. It is an amide bond, made when a carbon atom of one amino acid bonds to the nitrogen atom of the next, with one molecule of water lost (Hanazono et al., Science Advances, 2022). Peptide bonds run along the backbone of every peptide and protein, one between each pair of neighbouring amino acids. The carbon–nitrogen bond has partial double-bond character, which holds the atoms around it close to one flat plane and makes the bond resistant to hydrolysis, breakage by water (Hanazono et al., 2022).

The amide hydrogen and oxygen atoms of the bond often act as hydrogen-bond donors and acceptors, the weak links that help proteins hold their folds (Hanazono et al., 2022). That is why the bond matters beyond joining the chain: its shape and its hydrogen-bonding atoms set how a peptide can fold.

Flat is an idealisation. In a neutron structure of one iron-sulfur protein solved at 1.2 ångströms, Hanazono and colleagues found that the amide hydrogen nuclei deviate from the peptide plane, with the flatness of the H–N–C=O group depending on how far the nitrogen atom pyramidalises, or bends out of the plane, a finding that describes that structure (Hanazono et al., 2022). Textbook drawings show the ideal case, and real structures show small distortions.

How are peptide bonds formed?

Peptide bonds form by a dehydration reaction between amino acids, and in living cells the ribosome carries it out. In atomic structures of the large ribosomal subunit of one archaeon, Haloarcula marismortui, with two substrate analogs, the active site was made entirely of RNA, with no protein side-chain atom within about 18 ångströms of the bond being made, which showed that the ribosome is a ribozyme, an RNA catalyst (Nissen et al., Science, 2000). A 2011 laboratory kinetic isotope study of the transition state, which used a peptidyl-transfer RNA mimic and was a chemistry experiment and not a clinical study, found a two-step mechanism in which a tetrahedral intermediate forms and then breaks down in a separate fast step (Hiller et al., Nature, 2011).

Chemists make the same bond without a ribosome. Solid-phase peptide synthesis, described by Merrifield, builds a chain one amino acid at a time on a solid support, and its Fmoc version is now the method of choice for peptide synthesis (Merrifield, Methods in Enzymology, 1997; Behrendt et al., Journal of Peptide Science, 2016). Synthetic peptides get their own section later on this page.

What is the formula of a peptide bond?

The peptide bond is written as –C(=O)–NH–, a carbonyl carbon bonded to an amide nitrogen. The reaction that makes it can be written R1–COOH + H2N–R2 → R1–C(=O)–NH–R2 + H2O, where R1 and R2 stand for the rest of each amino acid and water is the by-product. Hanazono et al. describe peptide bonds as amide bonds generated by the dehydration reaction between α-amino acids (Science Advances, 2022). The partial double-bond character of the carbon–nitrogen bond comes from resonance, a sharing of electrons between the nitrogen and oxygen atoms (Hanazono et al., 2022).

A peptide's molecular formula is a different thing: the full atom count of the whole molecule. Dihexa, a modified peptide-like molecule, has the formula C27H44N4O5 and a molecular weight of 504.7, and glutathione, a tripeptide, has C10H17N3O6S and 307.33 (PubChem record for dihexa and for glutathione, checked 14 October 2026). Laboratories confirm that a measured mass matches the expected formula, a check explained in how mass spectrometry confirms peptide identity.

How is the number of peptide bonds in a peptide calculated?

A linear peptide of n amino acids has n minus 1 peptide bonds, because each bond joins one neighbouring pair. C-peptide, a 31-amino-acid peptide cut from proinsulin, has 30 peptide bonds along its chain (Leighton et al., Diabetes Therapy, 2017). A head-to-tail cyclic peptide, a ring in which the two ends of the chain join, has one bond for every amino acid; a ring closed through a side chain or a disulfide bridge is a different case. Each bond forms with the loss of one water molecule, so the mass of a linear chain is the sum of its amino acid masses minus one water for every bond (Hanazono et al., 2022). A tripeptide therefore has two bonds.

Counting bonds is arithmetic, not a research finding, so no study is needed for the rule itself. What studies add is the exceptions: bonds that depart from the flat ideal (Hanazono et al., 2022) and molecules in which a non-amino-acid piece is attached to the chain, as in the Dihexa record described below.

What foods have peptide bonds?

Every food that contains protein has peptide bonds, because proteins are long chains of amino acids joined by them. Digestion breaks those bonds in food proteins, and the intestine absorbs the resulting di- and tripeptides through the PEPT1 transporter, which accepts neither free amino acids nor peptides of four or more amino acids (Daniel, Annual Review of Physiology, 2004). Food-derived bioactive peptides, short sequences hidden in food proteins and released by digestion or by processing, are an active research area, though well-designed clinical trials to support health claims remain scarce (Daliri et al., Foods, 2017).

The sources checked rank no food by peptide-bond count, and the count of bonds is not a measure of food quality. Protein-rich foods simply carry more of them.

How does the body get peptides naturally?

The body gets peptides in two ways: it makes them, and it frees them from the protein in food. Cells build chains on ribosomes by forming peptide bonds one after another (Nissen et al., 2000). Some peptides are cut from larger precursors, as C-peptide is from proinsulin, which yields insulin and C-peptide in equal amounts (Leighton et al., 2017). Digestion of dietary protein produces short peptides that the intestine absorbs (Daniel, 2004). These sources describe normal biology, and none of them describes any way of changing the level of a particular peptide.

The question how to get peptides naturally is therefore answered here as biology, not as advice. The TGA adds a separate statement about products: it has not assessed unapproved peptide products for safety, quality or effectiveness (TGA, 23 September 2026).

Is Dihexa a peptide?

Dihexa is a peptide-like molecule, not a standard peptide. Its PubChem record gives the formula C27H44N4O5, a molecular weight of 504.7 and a name that describes a hexanoyl group, tyrosine and isoleucine joined to an aminohexanoic amide, so only two of its units are standard amino acids (PubChem record for dihexa, checked 14 October 2026). Dihexa was designed as an analogue of angiotensin IV, a natural hexapeptide in the brain (Sun et al., Brain Sciences, 2021). Such molecules are often called modified peptides.

Evidence limits come first. The sources checked contain animal work and no human trial. In APP/PS1 mice, a mouse model of Alzheimer's disease, with no mouse count in the abstract, Sun et al. (2021) reported better spatial-learning test results and lower brain inflammatory signals after Dihexa, findings that do not establish an effect in humans (Sun et al., Brain Sciences, 2021). No human trial or approval record for Dihexa was found in the sources checked.

What are synthetic peptides?

Synthetic peptides are peptides made in a laboratory or factory by chemistry or biotechnology, rather than taken from a living source. Chemical synthesis builds the chain amino acid by amino acid on a solid support, a method Merrifield described and one that is routine for chains under about 50 amino acids (Merrifield, Methods in Enzymology, 1997; Wang et al., 2022). Fmoc solid-phase synthesis is the method of choice today, with therapeutic peptides made this way in multiton quantities (Behrendt et al., 2016).

Synthetic does not mean unregulated or approved. Wang et al. note that some peptide drugs, such as oxytocin and teriparatide, use chemical synthesis to make their active ingredient (Wang et al., 2022), and many unapproved products sold online are synthetic too. The TGA says unapproved peptide products have not been assessed for safety, quality or effectiveness, so there is no assurance that they are correctly labelled, sterile or made to appropriate quality standards (TGA, 23 September 2026).

How the quality of a synthetic peptide is checked is a separate subject, covered in how HPLC purity is measured for peptides. The peptide comparison guide also sets natural and synthetic peptides side by side.

What are synthetic peptide vaccines?

Synthetic peptide vaccines use short, laboratory-made pieces of a germ's proteins, called epitopes, to train the immune system, in place of the whole organism. Classic vaccines use live or weakened microorganisms or parts of them, and immunologically unneeded components or impurities in such vaccines can cause problems that fully synthetic peptide vaccines might avoid (Skwarczynski and Toth, Chemical Science, 2016). The trade-off is that a minimal epitope alone gives a poor immune response, so peptide vaccines need an adjuvant, a substance that strengthens the response, and adjuvants bring their own difficulties (Skwarczynski and Toth, 2016).

The same review covers epitope design, adjuvants and nanoparticle-based approaches, and sets out the disadvantages as well as the advantages (Skwarczynski and Toth, 2016). Wang et al. describe peptide vaccine candidates built around epitopes of the SARS-CoV-2 spike protein, picked with immunoinformatics, a computer method for finding epitopes (Wang et al., 2022). Those are research designs. The sources checked for this page do not establish which peptide vaccines are approved, so none is named as approved.