Peptide and protein structure: primary, secondary, tertiary and quaternary

Protein structure is described at four levels: primary (the amino acid sequence), secondary (local helices and sheets), tertiary (the 3D fold of one chain) and quaternary (how several chains fit together). Long chains fold into precise three-dimensional forms, held together by many weak attractions, while short peptides usually don't. This guide explains each level, the forces involved, what happens when a structure is lost, and how structure is measured.

For the building blocks, see the 20 amino acids and what is a peptide. Terms are defined in the glossary.

The four levels of protein structure

Each level depends on the one before it. The sequence decides which local shapes are possible, and those shapes pack together into the overall fold.

Primary structure of a protein

The primary structure of a protein or peptide is the sequence: the order of amino acids from the N-terminus to the C-terminus, written in one- or three-letter codes. It also includes the positions of any disulfide bonds between cysteines and any modifications, such as a C-terminal amide.

Primary structure is the only level that is fixed by covalent bonds, so it is the only one that survives harsh treatment. It is what mass spectrometry confirms and what a sequence on a product page, such as the Semax heptapeptide sequence, describes. See amino acid codes and how to read a peptide sequence.

Secondary structure of proteins

Secondary structure describes regular local shapes of the backbone. They are held in place by hydrogen bonds between the C=O of one peptide bond and the N–H of another. See chemical bonds and polarity for how hydrogen bonds work.

The alpha helix

In an alpha helix, the backbone coils into a right-handed spiral, like a spring. In short: 3.6 residues per turn, a rise of 0.54 nm per turn, and a hydrogen bond from each C=O to the N–H four residues along.

• Each turn holds about 3.6 amino acids and rises about 0.54 nm.
• Each C=O hydrogen-bonds to the N–H four residues further along, so every backbone group inside the helix is bonded.
• The side chains point outwards from the spiral.

Proline breaks or kinks a helix, because its ring locks the backbone and it has no N–H to donate. Glycine is also uncommon in helices, because it is too flexible. See the 20 amino acids.

The beta sheet

In a beta sheet, stretches of the chain lie almost fully extended, side by side. Hydrogen bonds form between neighbouring strands rather than within one.

• Antiparallel sheets have neighbouring strands running in opposite directions.
• Parallel sheets have them running the same way.
• The side chains point alternately above and below the sheet.

Beta strands can come from distant parts of the same chain or from different chains. Sheets that stack between many chains are one way peptides aggregate.

Turns and disordered regions

Turns reverse the direction of the chain, often in just four residues, and frequently contain proline or glycine. Regions with no regular shape are called random coil or disordered. They aren't truly random, but they don't hold one fixed form.

Tertiary structure

The tertiary structure is the complete three-dimensional fold of a single chain: how its helices, sheets and loops pack against each other. It is held together by interactions between side chains:

• Hydrophobic effect: non-polar side chains cluster in the core, away from water. This is usually the main driving force of folding.
• Hydrogen bonds between side chains and with the backbone.
• Salt bridges between positive side chains, such as lysine, and negative ones, such as aspartate.
• Van der Waals forces between closely packed atoms.
• Disulfide bonds between cysteines, the only covalent links in this list.

Van der Waals forces

Van der Waals forces are weak attractions between all atoms that come close together. They arise because the electrons around an atom are always moving, which creates brief, shifting charges that attract neighbouring atoms. Each one is tiny, usually under a few kilojoules per mole, but a folded protein packs thousands of atoms tightly together, and the total is large.

The same forces help explain retention on a reversed-phase HPLC column. The column is packed with silica carrying C18 chains, straight chains of 18 carbon atoms. Non-polar side chains, such as those of leucine, isoleucine and phenylalanine, lie against those chains and are held by van der Waals contact, alongside the hydrophobic effect. Compare two peptides of the same length: one rich in leucine and one rich in serine. The leucine-rich peptide has more non-polar surface touching the C18 chains, so it usually stays on the column longer and comes off later as the organic solvent increases.

Quaternary structure

Some proteins are made of more than one chain. Quaternary structure describes how the separate chains, called subunits, fit together. A protein made of two identical chains is a dimer; four makes a tetramer. The same forces that hold a single chain together hold the subunits to each other, sometimes with disulfide bonds between chains.

Quaternary structure belongs to large proteins. Short synthetic peptides don't have it in a defined sense, although they can aggregate.

Protein folding

Protein folding is the process by which a chain finds its structure. Anfinsen's work on the enzyme ribonuclease showed that an unfolded chain can refold correctly on its own, so the sequence carries the information for the fold (Anfinsen, Science 1973). Folding is driven mainly by burying hydrophobic side chains away from water, with hydrogen bonds and other interactions fixing the details.

Folding can go wrong. Chains that misfold, or that are crowded at high concentration, can stick together into aggregates. In the lab, aggregation shows up as cloudiness, gels or lost material. See peptide solubility.

Denaturation

Denaturation is the loss of secondary, tertiary and quaternary structure. The chain unfolds but its sequence stays intact, because no peptide bonds are broken. Common causes in the lab:

Denaturation is sometimes reversible: removing the denaturant lets some proteins refold. Often it isn't, especially if the unfolded chains aggregate. SDS-PAGE relies on denaturing proteins on purpose, and reducing agents break disulfides before it.

Denaturation is different from degradation, which breaks covalent bonds and changes the molecule itself. See hydrolysis.

Do short peptides have a structure?

Most of the levels above apply to proteins and longer peptides. Short peptides behave differently.

• Little fixed structure in water. A short peptide, of a few to a few dozen residues, usually doesn't have enough internal interactions to hold one shape. It moves between many shapes, and is described as flexible or disordered.
• Structure can depend on the solvent. Some short peptides form helices in solvents such as trifluoroethanol that strengthen hydrogen bonds, but not in water.
• Disulfides and rings add constraint. A disulfide bond or a cyclic backbone limits the shapes a short chain can take.

For short research peptides, the primary structure, confirmed by mass spectrometry and checked for purity by HPLC, is the structure that matters for identity.

How structure is measured

Circular dichroism

Circular dichroism measures how differently a sample absorbs left- and right-circularly polarised light. Because amino acids and the folded backbone are chiral, each type of secondary structure gives a recognisable curve in the far ultraviolet, from about 190 to 250 nm (Greenfield, Nat. Protoc. 2006):

Band positions are from Greenfield, Nat. Protoc. 2006.

CD is a research method rather than a routine quality test. See NMR, FTIR and other methods and light and the electromagnetic spectrum.

Related guides

For the bonds and forces behind structure, see chemical bonds and polarity. For the chemical groups in each side chain, see functional groups in peptides. For the chain itself, see what is a polypeptide. For which familiar compounds are proteins rather than peptides (collagen is a protein), see is it a peptide?