A peptide is a chain of amino acids joined by peptide bonds — and the only thing separating a peptide from a protein is length, by convention rather than by chemistry. The boundary is usually drawn around fifty residues, it is not enforced by anything physical, and knowing where the conventions are arbitrary is what makes the rest of the terminology usable.
The bond, and the direction
Every amino acid has an amino group at one end, a carboxyl group at the other, and a side chain that makes it distinctive. A peptide bond forms when the carboxyl of one condenses with the amino group of the next, releasing water. Repeat, and a chain results.
The chain has direction. One end retains a free amino group — the N-terminus — and the other a free carboxyl, the C-terminus. Sequences are written N to C by universal convention, so the first residue named is the one at the amino end. Direction matters: the same residues in reverse order form a different molecule with the same mass, which is one reason a mass measurement confirms identity without proving sequence.
| Term | Typical length | Notes |
|---|---|---|
| Amino acid | 1 | The monomer |
| Dipeptide, tripeptide | 2–3 | Named by count |
| Oligopeptide | ~2–20 | Loosely defined |
| Polypeptide | ~20–50 | Loosely defined |
| Protein | >~50 | Often implies a folded structure |
Where exactly is the line between a peptide and a protein?
There is not one, and treating the fifty-residue figure as a rule causes more confusion than it removes. Some definitions turn on length, others on whether the molecule adopts a stable folded structure, others on whether it was made by a ribosome. Insulin is 51 residues and is called both. The useful question is never which word applies but whether the molecule has a defined three-dimensional structure — because that determines whether it can be denatured, which is a real physical property rather than a naming convention.
Does a peptide have a shape?
Short ones mostly do not, and that is a genuine difference from proteins rather than a lesser version of one. A peptide of a few residues is flexible in solution, sampling many conformations rather than holding one. Structure generally requires enough chain to form stabilising interactions, so below roughly twenty residues a defined fold is the exception. This has practical consequences: an unstructured peptide cannot be denatured in the usual sense, it has no folded core protecting it from proteases, and it presents its residues to a surface freely — which is why short peptides stick to tubes more readily than folded proteins do. Cyclising a short sequence is the standard way to give it a fixed shape, with consequences of its own.
Twenty side chains, four behaviours
The backbone is identical everywhere along the chain. Everything distinctive comes from the side chains, and for practical purposes they sort into four groups.
Charged. Lysine, arginine and histidine carry positive charge at neutral pH; aspartate and glutamate carry negative. These determine how a peptide behaves in a solvent and how many counter-ions the vial holds.
Polar. Serine, threonine, asparagine, glutamine, tyrosine and cysteine. Asparagine and glutamine are the residues that deamidate — a common degradation route that changes mass by one dalton and is easy to miss.
Hydrophobic. Leucine, isoleucine, valine, phenylalanine, methionine, alanine, tryptophan. A sequence dominated by these will not dissolve in water at a useful concentration.
Structural oddities. Glycine has no side chain and allows the backbone to bend freely. Proline's side chain loops back to the backbone, constraining it — which is why proline-rich stretches resist peptidases. Cysteine's thiol forms disulphide bridges, and is the reason a free cysteine makes a peptide a poor candidate for co-formulation. Glutathione is the textbook case.
Why do some residues appear in a sequence that are not among the twenty?
Because the molecule was synthesised rather than translated. A ribosome is limited to the standard set; solid-phase synthesis is not, so a designed peptide can incorporate D-amino acids, 2-aminoisobutyric acid, or entirely artificial residues. These are almost always there to defeat a specific protease, and their presence is a reliable signal that the molecule was engineered for stability rather than copied from nature.
What is the difference between a fragment and the protein it came from?
Everything except a shared stretch of sequence, and the distinction is regularly blurred in this field. A fragment is a short piece taken from a longer molecule; it lacks the parent's other domains and its folded structure, and it is not obliged to reproduce its behaviour. TB-500 is a seven-residue piece of a 43-residue protein and is routinely called thymosin β4. Semax comes from an ACTH fragment and does not act like ACTH. AOD-9604 is fifteen residues of growth hormone, and cannot present the binding surfaces the whole hormone uses. When reading a citation, establishing whether it studied the fragment or the parent is a first-order question, not a technicality.
How is a peptide's molecular weight calculated?
Sum the residue masses and subtract one water molecule per bond formed — or read it from the reference table — — a ten-residue peptide loses nine. Then account for modifications: amidation removes about one unit, acetylation adds 42, an acylation chain several hundred. This is why a calculated mass from a bare sequence often disagrees slightly with the figure on a certificate, and why that disagreement is usually the modification rather than an error. It also has nothing to do with how much of the powder is peptide, which is a separate question answered by net peptide content.
All products referenced here are supplied for laboratory and research use only. They are not drugs, foods, supplements or cosmetics, and are not for human or veterinary use.
