Peptide vs Polypeptide vs Protein: What Is a Polypeptide?

There is no single official cut-off, so it helps to know which convention a source is using. MeSH draws its oligopeptide–polypeptide line at about 12 to 13 amino acids. Pharmacology reviews commonly define therapeutic peptides as chains of fewer than 50 amino acids, one of them adding a molecular weight below 10 kDa.^[1][4][5]

Regulators draw their own line. For its biologics framework, the US FDA treats polymers of 40 or fewer amino acids as peptides rather than proteins. Protein products such as insulin and human growth hormone, historically approved as ordinary new drug applications, were deemed biologics licences on 23 March 2020.^[6]

So the same molecule can be a peptide in one document and a protein in another. Tesamorelin is a good example: it is widely called a peptide, yet its US prescribing information describes the 44-amino-acid sequence of human growth-hormone-releasing factor plus a hexenoyl group — above the FDA's 40-residue peptide line.^[13][6]

Not quite. MeSH defines proteins as linear polypeptides made on ribosomes that may then be modified, cross-linked, cleaved or assembled into complexes of several subunits, and it stresses that the amino-acid sequence determines the shape the chain takes as it folds — and with it the protein's function.^[2]

How a sequence dictates a folded structure has been one of biology's central questions for half a century. Computer simulations can now fold some small proteins accurately, and structure prediction has become far more successful than early researchers thought possible.^[7]

Many proteins are built from more than one chain. Haemoglobin is a tetramer of two alpha- and two beta-globin chains, each carrying a haem group. Insulin is much smaller but also has two chains: an A-chain of 21 amino acids and a B-chain of 30.^[8][9]

Insulin also shows how one chain can become a two-chain protein. It is made as a larger precursor, named proinsulin in 1967, which is processed into the mature hormone; recombinant production follows a similar route, refolding the precursor and converting it into insulin with two enzymes.^[10][9]

Epitalon, a research tetrapeptide, is four amino acids long (Ala-Glu-Asp-Gly). Semax and selank are heptapeptides of seven. By the MeSH convention all three are oligopeptides, a subset of peptides.^[11][12][3]

Tesamorelin (44 amino acids plus a hexenoyl group) and insulin (51 amino acids in two chains) sit in the zone where conventions disagree. Recombinant human growth hormone, somatropin, is unambiguously a protein: its US prescribing information describes 191 amino acids and a molecular weight of 22,124 daltons, with a sequence identical to pituitary growth hormone.^[13][9][14]

Peptides are also cut from proteins. Most bioactive peptides — peptide neurotransmitters, neuropeptides and peptide hormones — are generated by proteolytic cleavage of larger precursor proteins, and the breakdown of haemoglobin yields peptides with biological activity of their own.^[8]

Short peptides are usually made chemically, one amino acid at a time, on a solid support. A 2016 review describes Fmoc solid-phase peptide synthesis as the method of choice, with multiton production of therapeutic peptides driving down the cost of the building blocks. Tesamorelin's prescribing information, for instance, describes it as produced synthetically.^[15][13]

Large proteins are usually made biologically. In 1979, synthetic genes for the insulin A- and B-chains were expressed separately in E. coli, and the purified chains were then mixed, reduced and reoxidised until insulin could be detected. Somatropin is likewise produced by recombinant DNA technology in E. coli.^[16][14]

Biological production brings its own folding problem. Recombinant proinsulin expressed in E. coli accumulates in inclusion bodies that have to be denatured and refolded before enzymes convert it into mature insulin.^[9]

Size changes how molecules behave. Unmodified peptides usually undergo extensive proteolytic cleavage, which gives them short half-lives, whereas proteins depend on a folded structure that can be lost and sometimes has to be restored by refolding.^[4][9]

It also changes how identity and purity are checked. An FDA laboratory showed that liquid chromatography with high-resolution mass spectrometry can confirm a peptide's sequence, determine its amino-acid composition and quantify impurities below 0.1% in a single run — the kind of evidence that belongs behind a peptide certificate of analysis.^[17]