How many amino acids are in insulin?

Mature human insulin contains 51 amino acids, split between two chains: an A chain of 21 residues and a B chain of 30. Three disulfide bonds hold the molecule together, two linking the chains and one inside the A chain. The other numbers that turn up in textbooks and quiz answers, 110 and 86, describe insulin's precursors before a beta cell trims them into the hormone.

Both human chain sequences are below, ready to paste into a tool such as Amino acid composition, followed by the precursor and the few changes that make insulin lispro, aspart and glargine different from the hormone the pancreas makes.

How many amino acids are in insulin?

One molecule of mature human insulin contains 51 amino acid residues, 21 in the A chain and 30 in the B chain.^[1][2] A residue is what remains of an amino acid once it has been joined into a chain, so for counts like this the two words mean the same thing.

Insulin is not one continuous chain. Positions are numbered from the start of each chain, so A7 is the seventh residue of the A chain. Disulfide bonds between cysteines A7 and B7 and between A20 and B19 join the two chains, and a third links A6 to A11 within the A chain.^[3][2] The molecule therefore has 49 peptide bonds, 20 in the A chain and 29 in the B chain. The disulfide bonds connect side chains and do not add to that count.

The 51 positions use 17 of the 20 standard amino acids; human insulin contains no aspartic acid, methionine or tryptophan.^[3] Its molecular weight is 5,808 Da.^[4] We calculated 5,814 Da for the two chains as free peptides; each disulfide bond releases two hydrogen atoms, which brings the total back to 5,808 Da, a correction the Protein molecular weight calculator can apply.

This count describes one insulin molecule. Beta cells store insulin as hexamers, three dimers held together by zinc ions, but each unit in the hexamer still has 51 residues.^[1]

Human insulin amino acid sequence

UniProt entry P01308 records the human sequence as part of the 110-residue precursor encoded by the INS gene, with the B chain at residues 25 to 54 and the A chain at residues 90 to 110.^[3] Cut out of that precursor, the two chains read as follows in one-letter code:

>Human insulin A chain, 21 residues (UniProt P01308, 90-110)
GIVEQCCTSICSLYQLENYCN
>Human insulin B chain, 30 residues (UniProt P01308, 25-54)
FVNQHLCGSHLVEALYLVCGERGFFYTPKT

In the three-letter code many textbooks use, the same chains are:

A chain: Gly-Ile-Val-Glu-Gln-Cys-Cys-Thr-Ser-Ile-Cys-Ser-Leu-Tyr-Gln-Leu-Glu-Asn-Tyr-Cys-Asn
B chain: Phe-Val-Asn-Gln-His-Leu-Cys-Gly-Ser-His-Leu-Val-Glu-Ala-Leu-Tyr-Leu-Val-Cys-Gly-Glu-Arg-Gly-Phe-Phe-Tyr-Thr-Pro-Lys-Thr

The six cysteines (C) sit at A6, A7, A11, A20, B7 and B19, the pairs behind the three disulfide bonds. Keep the chains as separate records when you analyze them. Insulin is two peptides, and a single 51-letter string would describe a chain that never exists in the cell.

Why do some sources say insulin has 110 amino acids?

The 110-amino-acid count describes preproinsulin, the single chain that ribosomes translate from insulin mRNA, while 51 describes the mature hormone.^[1][2] The difference is removed in two steps.

The first 24 residues form a signal peptide that sends preproinsulin into the endoplasmic reticulum, where it is cut off. What remains is 86-residue proinsulin, a single chain in which the C-peptide connects the future B and A chains while the three disulfide bonds form.^[2][1] In secretory granules, enzymes then cut out the 31-residue C-peptide and the pair of basic residues on each side of it, leaving the two chains of mature insulin.^[1]

The arithmetic follows from the table. Removing the 24-residue signal peptide leaves proinsulin's 86 residues, and removing the 31-residue C-peptide and the four basic residues from proinsulin leaves 51.

How do insulin analogs and animal insulins differ?

Most therapeutic insulins keep human insulin's 51-residue framework and change one to three positions. The exceptions in length are insulin glargine, with 53 residues, and insulin degludec, with 50 residues plus a fatty-acid side chain.

The changes target how insulin behaves after injection. In a vial, insulin assembles into zinc-bound hexamers, and that self-association slows its absorption from under the skin.

Brange and colleagues showed in 1988 that single amino acid substitutions could keep insulin essentially monomeric at pharmaceutical concentrations, and those insulins were absorbed two to three times faster than the rapid-acting insulins of the time.^[5] The rapid-acting analogs lispro, aspart and glulisine change residues at B28, B29 or B3 without altering either chain's length.^[4][6][7]

Long-acting analogs go the other way. Glargine replaces asparagine A21 with glycine, adds two arginines to the end of the B chain and is formulated at about pH 4; after injection it is absorbed slowly, giving a fairly constant insulin level over 24 hours.^[8] Degludec drops threonine B30 and carries a fatty-acid side chain on lysine B29, which makes it form multi-hexamers under the skin that release the hormone slowly.^[9]

The animal insulins show how little the sequence varies between pigs, cows and people. Pig insulin differs from human insulin only at B30, and cow insulin at B30, A8 and A10.^[10][11] Cow insulin was also the first protein to have its complete amino acid sequence worked out: Frederick Sanger's group finished it in 1955 by locating its three disulfide bonds.^[12]

Which number is right therefore depends on the molecule. The INS gene encodes 110 residues, the folding precursor in the endoplasmic reticulum has 86, the hormone a beta cell secretes has 51, and the analogs in a pharmacy range from 50 to 53. Through all of them the A chain stays at 21 residues.