The Peptide Bond

How the bond forms

Chemically, a peptide bond is a secondary amide. It links the carboxyl carbon of one amino acid to the α-amino nitrogen of the next, and the overall reaction is a condensation: one molecule of water is lost for every bond formed.

This is the reason peptide masses are built from residue masses, each one an amino acid minus water, as explained in the molecular weight guide. It also defines the direction of every chain. The end with the free amine is the N-terminus, the end with the free carboxyl is the C-terminus, and sequences are always written from N to C.

Simply mixing two amino acids in water does not produce a peptide. In aqueous solution the equilibrium lies strongly on the side of hydrolysis, and at neutral pH the reacting groups are present as an ammonium and a carboxylate, neither of which is reactive in the right way. Every route to peptide bonds therefore starts by activating the carboxyl group:

• Ribosomes use aminoacyl-tRNAs, in which the carboxyl group is activated as an ester. The peptidyl transferase centre, made of ribosomal RNA, positions the amine to attack it.
• Nonribosomal peptide synthetases in bacteria and fungi activate amino acids as adenylates and thioesters. They build many antibiotics and immunosuppressants, including vancomycin, cyclosporin and gramicidin S, and they can use D-amino acids and other non-standard building blocks.
• Chemical synthesis activates the carboxyl group with coupling reagents and protects every other reactive group, as described in the guide to solid-phase peptide synthesis.

Resonance and planarity

The lone pair on the amide nitrogen is not localised. It is shared with the adjacent carbonyl group, and the bond is best described as a hybrid of two resonance forms:

The real structure lies between the two, with the C–N bond carrying roughly 40% double-bond character. This has three measurable consequences.

The C–N bond is short. At about 1.33 Å it lies between a typical C–N single bond (1.47 Å) and a C=N double bond (about 1.27 Å). The neighbouring N–Cα and Cα–C bonds, which have no such character, are ordinary single bonds of 1.46 and 1.52 Å.

The bond cannot rotate freely. Twisting it out of plane breaks the overlap that stabilises it. The barrier is around 80 kJ/mol, high enough that the bond behaves as rigid at room temperature, but low enough that cis and trans forms can interconvert slowly.

Six atoms lie in one plane. The Cα of the first residue, the carbonyl C and O, the amide N and H, and the Cα of the next residue are all coplanar. Linus Pauling and Robert Corey established this from crystal structures of small peptides, and the rigid amide plane was the key constraint that let them predict the α-helix and β-sheet in 1951.

Resonance also makes the amide bond strongly polar. The oxygen carries a partial negative charge and the N–H a partial positive one, which makes the carbonyl oxygen a good hydrogen-bond acceptor and the N–H a good donor. These backbone hydrogen bonds hold together every α-helix and β-sheet.

Cis and trans peptide bonds

Because the amide bond is planar, the two Cα atoms can sit on opposite sides of the C–N bond (trans, ω ≈ 180°) or on the same side (cis, ω ≈ 0°). For most residue pairs the cis form puts the two side chains and backbone atoms into a steric clash, so trans is strongly favoured. A survey of protein crystal structures (Jabs, Weiss and Hilgenfeld, 1999) found only about 0.03% of non-proline peptide bonds in the cis form.

Proline is the exception. Its side chain loops back and bonds to its own backbone nitrogen, so the nitrogen has no H and carries a carbon substituent on both sides. The cis and trans forms of the bond before a proline (the X–Pro bond) are then much closer in energy. In the same survey about 5% of X–Pro bonds were cis, and in short unstructured peptides the cis fraction is commonly 10–30%, depending on the neighbouring residues.

This has practical consequences:

• NMR spectra of proline-containing peptides often show doubled peaks. Cis and trans forms interconvert slowly, on a timescale of seconds to minutes, so both appear as separate species. This is not an impurity.
• Reversed-phase HPLC can occasionally show broadened or split peaks for the same reason, especially at low temperature.
• Protein folding can be slowed by proline isomerisation. Cells use peptidyl-prolyl isomerases, such as cyclophilins and FKBPs, to speed it up. The immunosuppressants cyclosporin and tacrolimus bind these enzymes.

Backbone flexibility: φ and ψ

If the peptide bond is rigid, all the flexibility of a peptide backbone must come from the two bonds on either side of each Cα atom. Rotation about N–Cα is described by the torsion angle φ (phi), and rotation about Cα–C by ψ (psi). Every residue in a chain can therefore be described by one pair of angles.

In 1963 G. N. Ramachandran and colleagues showed that most combinations of φ and ψ are impossible because atoms of neighbouring residues would collide. The allowed combinations form a few well-defined regions of what is now called the Ramachandran plot:

Two residues behave differently. Glycine, with only a hydrogen as its side chain, can reach regions forbidden to other residues, which is why it is common in tight turns and loops. Proline has its φ angle locked near −65° by its ring and has no N–H to donate a hydrogen bond, so it breaks α-helices and β-sheets except at their ends.

A bond that is unstable but lasts

Hydrolysis of a peptide bond releases energy, so thermodynamically every peptide in water "wants" to fall apart. Kinetically, however, the bond is remarkably inert. Resonance makes the carbonyl carbon a poor target for attack by water, and the uncatalysed reaction at neutral pH and room temperature is extremely slow: estimated half-lives range from years to centuries depending on the bond and the conditions.

This combination of thermodynamic instability and kinetic stability is what biology needs. Proteins persist for as long as they are wanted, yet a protease can cleave a specific bond in a fraction of a second when required. The same principle explains why lyophilised peptides can be stored for years, as covered in the guide to peptide stability and storage.

Breaking peptide bonds deliberately

Controlled cleavage is central to peptide analysis. The main methods differ in how selective they are:

Complete acid hydrolysis is not perfectly clean. It destroys tryptophan, partly degrades serine and threonine, and converts asparagine and glutamine into aspartic and glutamic acid. Amino acid analysis therefore reports Asx and Glx rather than the individual residues, and tryptophan must be determined separately.

Isopeptide bonds

An amide bond that involves a side-chain group instead of the α-amino or α-carboxyl group is called an isopeptide bond. Several important molecules depend on them:

• Glutathione (γ-Glu–Cys–Gly) links glutamate through its side-chain γ-carboxyl. This protects it from most peptidases, which recognise only α-linked bonds.
• Ubiquitin is attached to target proteins through an isopeptide bond between its C-terminal glycine and the ε-amino group of a lysine.
• Transglutaminases cross-link glutamine and lysine side chains in blood clots (factor XIIIa), in skin, and in the gluten network of bread dough.

Isopeptide bonds have the same mass and formula as ordinary peptide bonds, so a mass measurement alone cannot tell them apart. The MW calculator gives the correct mass for glutathione even though ECG is not an ordinary α-linked sequence.

Seeing the peptide bond: UV absorbance

The amide group absorbs ultraviolet light strongly below about 230 nm, with a strong π→π* band near 190 nm and a weaker n→π* band near 210–220 nm. Two everyday techniques rely on this:

• HPLC detection at 214–220 nm sees every peptide, whether or not it contains aromatic residues. Detection at 280 nm sees only peptides with Trp or Tyr.
• Circular dichroism in the far UV (roughly 190–250 nm) reports on secondary structure, because the amide groups of an α-helix or β-sheet absorb left- and right-circularly polarised light differently. An α-helix gives characteristic minima at 208 and 222 nm.

Frequently asked questions

References

• Pauling L, Corey RB, Branson HR (1951) The structure of proteins: two hydrogen-bonded helical configurations of the polypeptide chain. PNAS 37:205–211.
• Ramachandran GN, Ramakrishnan C, Sasisekharan V (1963) Stereochemistry of polypeptide chain configurations. Journal of Molecular Biology 7:95–99.
• Jabs A, Weiss MS, Hilgenfeld R (1999) Non-proline cis peptide bonds in proteins. Journal of Molecular Biology 286:291–304.
• Radzicka A, Wolfenden R (1996) Rates of uncatalyzed peptide bond hydrolysis in neutral solution and the transition state affinities of proteases. Journal of the American Chemical Society 118:6105–6109.

Related guides

Browse all 27 guides →