A peptide bond is the amide bond that joins two amino acids: the carboxyl carbon of one amino acid bonds to the amino nitrogen of the next, and a molecule of water is released. The bond has partial double-bond character, so the six atoms around it sit in one flat plane, and it almost always adopts the trans arrangement. It is also kinetically very stable, lasting for centuries in neutral water at room temperature unless an enzyme, strong acid or strong base breaks it.
Every peptide and protein backbone is a chain of these bonds. Their geometry decides which shapes a chain can take, and their stability decides how long a peptide survives on the bench. This article covers how the bond forms, why it is flat, the cis and trans forms, how it breaks, and the practical consequences for handling and analyzing peptides.
What a peptide bond is
An amino acid carries an amino group (–NH2) and a carboxylic acid group (–COOH) on the same carbon, the alpha carbon. When the carboxyl group of one amino acid condenses with the amino group of another, the product is an amide: –C(=O)–NH–. Chemists call this link an amide bond in general; when it connects two amino acids in a chain, it is called a peptide bond.
The reaction is a condensation, also called a dehydration, because the atoms of one water molecule (–OH from the acid and –H from the amine) leave. Two amino acids joined this way make a dipeptide. The new molecule still has a free amino group at one end and a free carboxyl group at the other, so the process can repeat, building a chain with a direction: the N-terminus at the start and the C-terminus at the end. The article on what peptides are covers how those chains are named and classified by length.
| Feature | Detail |
|---|---|
| Bond type | Secondary amide, –C(=O)–NH– (tertiary when the nitrogen belongs to proline) |
| Formed by | Condensation of a carboxyl group and an amino group, releasing water |
| Broken by | Hydrolysis: addition of water, catalyzed by proteases, strong acid or strong base |
| Geometry | Planar, with Cα, C, O, N, H and the next Cα in one plane |
| Usual configuration | Trans; cis is uncommon except before proline |
| Hydrogen bonding | N–H is a donor, C=O is an acceptor |
How peptide bonds form
In water, simply mixing two amino acids does not produce a dipeptide in any useful amount. The carboxyl group has to be activated first, turned into a better leaving group so the amine can attack it. Both living cells and chemists solve the problem the same way in principle, by activating the carboxyl group, but with very different machinery.
On the ribosome
In cells, each amino acid is first attached to a transfer RNA as an ester. The ribosome brings two such charged tRNAs together and catalyzes the transfer of the growing chain onto the incoming amino acid. The catalytic center of the ribosome turned out to be made of RNA, not protein. In 2000, Nissen and colleagues at Yale solved structures of the large ribosomal subunit bound to substrate analogs and reported that the analogs were contacted only by conserved ribosomal RNA, with no protein side-chain atoms closer than about 18 angstroms to the peptide bond being made. The ribosome is, in their words, a ribozyme.
In the laboratory
Chemical synthesis activates the carboxyl group with a coupling reagent. Early methods used carbodiimides such as N,N'-dicyclohexylcarbodiimide (DCC), often with an additive such as 1-hydroxybenzotriazole (HOBt) to limit racemization. Later reagents such as HBTU, HATU and TBTU give faster coupling. To make sure the right bond forms and no other, every reactive group that should not take part is blocked with a protecting group, and the alpha-amine is deprotected only when it is its turn to react. That approach, carried out on a resin in solid-phase peptide synthesis, is how research peptides are built one bond at a time.
Why the peptide bond is planar
The nitrogen of an amide carries a lone pair of electrons that is shared with the neighboring carbonyl group. The bond can be drawn in two resonance forms: one with a C=O double bond and a C–N single bond, the other with a C–O single bond (negative charge on oxygen) and a C=N double bond (positive charge on nitrogen). The real bond sits between the two. The C–N link is therefore shorter than an ordinary C–N single bond and cannot rotate freely.
Linus Pauling, Robert Corey and Herman Branson built that constraint into their 1951 models of the polypeptide chain. By assuming the amide group was planar, they predicted the alpha helix before any protein structure had been solved at atomic detail. The same assumption is still used in structure refinement. Later work has refined it: a 2011 study in PLOS ONE combining quantum-mechanical calculations with crystal structures found that small deviations from perfect planarity are common and depend mainly on the neighboring backbone angle psi.
Backbone angles: phi, psi and omega
Because the peptide bond itself is locked, a chain bends only at the two single bonds on either side of each alpha carbon. The three backbone torsion angles are:
| Angle | Bond | Freedom |
|---|---|---|
| Phi (φ) | N–Cα | Rotates; restricted by steric clashes |
| Psi (ψ) | Cα–C | Rotates; restricted by steric clashes |
| Omega (ω) | C–N, the peptide bond | Essentially fixed near 180° (trans) or near 0° (cis) |
In 1963 G. N. Ramachandran and colleagues worked out which combinations of phi and psi are allowed without atoms colliding. The resulting map, the Ramachandran plot, shows that only a few regions are open, and those regions correspond to the alpha helix, the beta strand and a handful of turn types. A peptide chain can take many shapes, but far fewer than it could if every backbone bond rotated freely.
Trans and cis peptide bonds
A planar bond can exist in two arrangements. In the trans form, the two alpha carbons lie on opposite sides of the C–N bond; in the cis form, they lie on the same side. Trans is strongly favored because the cis form puts the two alpha carbons and their side chains close together.
Proline is the exception. Its side chain loops back and bonds to its own nitrogen, so that nitrogen carries two carbon substituents and no hydrogen, and the steric cost of the cis form drops. A 2011 analysis of high-resolution protein structures in BMC Bioinformatics reports that 5.2% of bonds preceding proline (Xaa-Pro) are cis, compared with 0.03% of bonds preceding other residues. Interconversion between the two forms is slow, and cells have enzymes, peptidyl-prolyl isomerases, that speed it up.
For a synthetic peptide that contains proline, this means that the same molecule can exist as more than one conformer in solution. Proline-rich sequences can show broadened or split signals in NMR spectra, and occasionally a shoulder in HPLC, that reflect cis and trans populations rather than an impurity.
Hydrogen bonding and polarity
Each peptide bond carries a hydrogen-bond donor (the N–H) and an acceptor (the C=O). In a folded protein, these groups bond to each other in a regular pattern, which is what holds together alpha helices (each C=O bonded to the N–H four residues along) and beta sheets (bonds between neighboring strands). Proline's nitrogen has no hydrogen, so it cannot donate one, which is why proline often interrupts helices.
The same polarity makes the backbone interact strongly with water. A 2021 review of short peptides in Molecules noted that this interaction is one reason peptides cross lipid membranes poorly. In the lab, it is part of why short, charged peptides usually dissolve readily in aqueous buffers while long hydrophobic stretches may need an organic co-solvent.
How stable is a peptide bond?
Thermodynamically, hydrolysis of a peptide bond in water is favorable. Kinetically, it is extremely slow. Without a catalyst, at neutral pH and 25 °C, measured half-lives for peptide bond hydrolysis are in the range of hundreds of years; a 2018 review in Molecules cites a half-life of 267 years for amide bonds at pH 7 and 25 °C, similar to earlier values from Radzicka and Wolfenden. That combination, thermodynamically unstable but kinetically inert, is what lets cells build durable proteins and still break them down on demand with proteases.
Outside the cell, the bond breaks under three main conditions:
- Enzymes. Proteases and peptidases cut peptide bonds, many of them with strong preferences for particular residues on either side. Trypsin, for example, cuts after lysine and arginine.
- Strong acid. Heating a peptide in concentrated hydrochloric acid for many hours breaks it down to its free amino acids, which is how amino acid analysis works.
- Strong base. Hydroxide attacks the carbonyl carbon directly, though side reactions such as racemization make this less useful analytically.
Some bonds are weaker than average. Bonds next to aspartic acid are a known soft spot: in solid-phase synthesis, the Asp-Gly sequence is the worst case for aspartimide formation, a side reaction in which the side chain of aspartic acid attacks the backbone. For the finished, freeze-dried peptide, moisture is the main driver of slow hydrolysis, which is why the article on lyophilized peptides spends so much time on residual water.
Variations on the standard bond
Not every amide bond in a peptide is a standard alpha-peptide bond, and not every link between residues is an amide.
| Variation | What differs | Example |
|---|---|---|
| Isopeptide bond | Uses a side-chain carboxyl or amine instead of the alpha group | Glutathione (gamma-glutamyl bond); lysine side-chain linkages |
| Tertiary amide | Nitrogen carries no hydrogen | Every bond before proline; N-methylated synthetic peptides |
| Cyclic backbone | Head-to-tail amide closes a ring | Cyclic research peptides |
| Disulfide bridge | S–S bond between two cysteines, not an amide | Oxytocin (positions 1 and 6) |
| C-terminal amide | The final carboxyl is an amide, not an acid | Oxytocin, many synthetic analogs |
Glutathione is a useful example. It is a tripeptide of glutamic acid, cysteine and glycine, but its first link runs through the gamma (side-chain) carboxyl of glutamic acid. That isopeptide bond is not recognized by most ordinary peptidases, and it is one reason glutathione is not made on the ribosome but by dedicated enzymes. Anhydrolabs supplies glutathione as a reagent for in-vitro work.
Why the peptide bond matters in the lab
Several routine laboratory techniques rely directly on the chemistry of the peptide bond.
HPLC detection. The amide group absorbs ultraviolet light in the far UV, so reversed-phase HPLC methods for peptides commonly read absorbance at around 210 to 220 nm, where every residue contributes and not only aromatic ones. Purity is then reported as the area of the principal peak relative to the total. The article on peptide purity explains how that number is calculated.
Mass spectrometry. In tandem mass spectrometry, peptide ions break mainly along their peptide bonds, producing ladders of fragments that can be read back into a sequence. This is how the identity of a synthetic peptide is confirmed beyond its intact mass, as described in how to read a peptide certificate of analysis.
Storage. Because water is required to break a peptide bond, keeping a peptide dry is the most important single step in keeping it intact. Anhydrolabs supplies its peptides as lyophilized powder in vacuum-sealed vials, and as 10-vial kits, for this reason. Once a stock solution is made, it should be stored cold, in aliquots, and used within the window the sequence allows.
