What is a peptide?

What is a peptide? A chain of amino acids, joined end to end

The question what is a peptide has a one-line answer: a peptide is a molecule built from amino acids joined in a line. Each amino acid is linked to the next by a peptide bond, and the order in which they are linked is called the sequence. Everything else on this page is vocabulary stacked on top of that.

Amino acids are the units. Each carries an amino group at one end, a carboxyl group at the other, and a side chain that makes it the amino acid it is. Twenty of them are encoded genetically and make up almost every natural peptide and protein. The side chains are where the chemical variety lives, because the backbone of every peptide is the same repeating pattern of nitrogen, carbon and carbonyl carbon.

Joining two amino acids gives a dipeptide. Three gives a tripeptide. Seven gives a heptapeptide, and fifteen a pentadecapeptide, the word that turns up on the specification for BPC-157. Past about ten the counting names stop being useful and the material is simply called a peptide, then a polypeptide, then a protein.

Where a peptide stops and a protein starts has no chemical answer, because there is no chemical difference. The convention most references use puts the line at about fifty residues, and it is a habit rather than a property of the molecule. A chain of forty-nine is not built differently from a chain of fifty-one.

One consequence matters for anyone reading a label. Because sequence defines the molecule, two compounds can share a trade name and not be the same substance. The clearest case in this catalogue is TB-500, a name used both for the seven-residue acetylated fragment Ac-LKKTETQ at 889.0 g/mol and for the full-length thymosin beta-4 chain, which is several times heavier. Molecular weight is the field that separates them in one glance.

The peptide bond, and why a residue weighs about 18 less than its amino acid

Answering what is a peptide chemically means answering what a peptide bond is. The bond that joins two amino acids is an amide bond, and in this context it is called a peptide bond. It forms between the carboxyl group of one amino acid and the amino group of the next: the carboxyl gives up an OH, the amino group gives up an H, and those three atoms leave together as a molecule of water. A reaction that joins two things and releases water is a condensation, and a peptide chain is a string of them.

That lost water is why the arithmetic on a specification works the way it does. A free amino acid has a known mass. The same amino acid sitting inside a chain has lost 18.02 of it, and is no longer called an amino acid at all: inside the chain it is a residue, and its residue mass is the free mass minus the mass of water.

Work it on the smallest possible case. Glycine is C2H5NO2 at 75.07 g/mol. Join two of them and the result is glycylglycine, a dipeptide: 75.07 plus 75.07 is 150.14, minus 18.02 for the water released gives 132.12 g/mol. One bond, one water, one subtraction.

Now do it on a real catalogue compound. GHK is glycine, histidine and lysine in that order. The free amino acids weigh 75.07, 155.15 and 146.19, which sum to 376.41. Three residues are joined by two peptide bonds, so two waters leave: 376.41 minus 36.04 gives 340.37 g/mol for the free tripeptide. The GHK-Cu copper peptide sold here is that tripeptide holding a copper(II) ion, and copper weighs 63.55, which is where the 403.94 g/mol on the specification comes from.

The general rule follows: a chain of n residues contains n minus 1 peptide bonds, so the mass of the peptide is the sum of the free amino acid masses minus n minus 1 waters. A 15-residue chain has lost fourteen waters, about 252 g/mol, before anything else is counted.

One property of the bond explains why a peptide is not a floppy string: it has partial double-bond character, which holds the atoms around it flat and stops them rotating.

Residues, the N-terminus and the C-terminus

A residue is what an amino acid becomes once it is part of a chain. The word matters because it is the unit everything else is counted in: a sequence is a list of residues, a chain length is a residue count, a deletion impurity is a chain missing one residue.

A chain has two different ends, and they are not interchangeable. At one end there is a free amino group, and that end is the N-terminus, N for nitrogen. At the other there is a free carboxyl group, and that is the C-terminus, C for carbon. Every residue in between has both its groups tied up in peptide bonds.

By universal convention a sequence is written and read left to right, from the N-terminus to the C-terminus. That is the direction used in every database, on every certificate and in every paper, and it is also the direction in which a ribosome builds a chain in a cell. Reversing a sequence gives a different molecule, not the same one written backwards.

Take the TB-500 sequence as it appears on the label: Ac-LKKTETQ. Strip the prefix for a moment and read the letters. L is leucine and sits at the N-terminus. Then K, lysine. Then K again. Then T, threonine. Then E, glutamic acid. Then T again. Then Q, glutamine, at the C-terminus. Seven letters, seven residues, six peptide bonds holding them together. Positions are numbered from the N-terminus too, so leucine is residue 1 and glutamine is residue 7, and when a fragment is named as a range, such as GRF(1-44), the numbers count the same way from the N-terminal end of the parent chain.

One-letter and three-letter codes, read on a real sequence

Every standard amino acid has two abbreviations. The three-letter code is readable and is used in prose and on labels where space allows: Gly, His, Lys. The one-letter code is a single capital and is used for anything longer than a few residues, because a 44-residue sequence in three-letter code is unreadable and in one-letter code is one line.

Most one-letter codes are the first letter of the name. G for glycine, L for leucine, P for proline. The exceptions exist because several names start with the same letter, so the set was assigned to avoid collisions: K for lysine, W for tryptophan, F for phenylalanine, Y for tyrosine, R for arginine, Q for glutamine and N for asparagine are the ones that catch people out. Here is the full set, with the catalogue sequences each letter appears in.

Now read a real sequence straight through. BPC-157 is written GEPPPGKPADDAGLV. That is glycine, glutamic acid, proline, proline, proline, glycine, lysine, proline, alanine, aspartic acid, aspartic acid, alanine, glycine, leucine, valine. Fifteen letters, fifteen residues, fourteen peptide bonds, and a run of three prolines near the N-terminus which is a structural feature rather than a typographical error. The formula is C62H98N16O22 and the mass is 1419.53 g/mol, and the BPC-157 peptide is the same molecule whichever of those three notations it is written in.

The same method reads MOTS-c, written MRWQEMGYIFYPRKLR: methionine, arginine, tryptophan, glutamine, glutamic acid, methionine, and so on to the final arginine. Sixteen residues, 2174.61 g/mol, every letter checkable against the table above. If a sequence printed on a certificate does not match the sequence on the product page character for character, the two documents describe different molecules, and that is worth catching before anything else on either page is read.

What is a peptide next to a protein and a single amino acid?

Three words get used as if they named a hierarchy of different things, and they are really one kind of thing at three sizes. An amino acid is a single unit. A peptide is a short chain of them. A protein is a long chain of them. The chemistry of the backbone is identical in all three cases, so the honest answer to what is a peptide against what is a protein is: length, by convention.

Size does change behaviour, which is why the vocabulary survives. A short peptide is usually flexible and often has no fixed fold at all. A long chain folds into a defined three-dimensional structure held together by its side chains, and that folded shape is what people mean by protein structure. The transition between the two is gradual.

The ladder below runs from a single amino acid to a chain that reference databases call a protein. Two rows make the point that the fifty-residue boundary is a convention: thymosin beta-4 is a 43-residue chain described as a protein, tesamorelin a 44-residue chain described as a peptide. One residue apart, different words, no chemical difference.

The masses are worth a second look. Residue count and mass are not interchangeable, because residues are not the same size. TB-500 has seven residues and weighs 889.0 g/mol, an average of about 127 per residue; MOTS-c has sixteen and weighs 2174.61, an average of about 136. The difference is in the side chains. A specification prints both numbers because neither implies the other.

Modifications on a label: Ac-, -NH2 and non-standard residues

A sequence of letters is often not the whole molecule. Peptides are routinely modified at one or both ends, and those modifications are printed as part of the sequence line because they change the compound, its mass and its formula.

Ac- at the front means the N-terminus is acetylated. An acetyl group, CH3CO, has replaced one hydrogen on the terminal amino group, adding 42.04 to the mass and removing the positive charge a free amino group carries. TB-500 is written Ac-LKKTETQ for exactly this reason: the seven letters are the chain and the Ac- is a real chemical change to its front end. A molecule with the same seven letters and a free N-terminus is a different compound with a different mass.

A trailing -NH2 means the C-terminus is amidated. The terminal carboxyl group, COOH, has been replaced by a carboxamide, CONH2, which removes the negative charge the free acid would carry. Ipamorelin is written Aib-His-D-2-Nal-D-Phe-Lys-NH2, and that -NH2 is the amidated end. So is the one on tesamorelin, written trans-3-hexenoyl-GRF(1-44)-NH2, where the front of the molecule carries an acyl group instead of an acetyl one.

Non-standard residues appear by name rather than by letter, because there is no one-letter code for them. Ipamorelin has three in a five-residue chain: Aib is alpha-aminoisobutyric acid, which is not one of the twenty; D-2-Nal is D-3-(2-naphthyl)alanine; and D-Phe is phenylalanine in its D form. Amino acids are chiral, and almost all of those in natural peptides are the L form. A D residue is the mirror image and makes a genuinely different molecule. One-letter code silently loses that distinction, which is why chains containing D residues are spelled out in three-letter code.

Metal ions are a modification of a different kind. GHK-Cu is the GHK tripeptide holding a copper(II) ion, and the copper is part of the substance the CAS number refers to. So read the whole sequence line, including everything before the first letter and after the last.

Natural or synthetic: what is a peptide made in a synthesiser?

Peptides occur throughout biology. Cells make them on ribosomes from a genetic template, and many more are produced by cutting longer proteins into fragments. Some compounds here have a natural parent: the MOTS-c peptide is encoded in mitochondrial DNA, and TB-500 corresponds to a fragment of the thymosin beta-4 protein. Others, such as ipamorelin, contain residues that do not occur in natural peptides at all.

The chemistry does not care where a molecule came from. A peptide assembled in a synthesiser with the same sequence, the same terminal modifications and the same stereochemistry as a natural one is the same molecule. Same formula, same molecular weight, same CAS number, same behaviour in an instrument. There is no test that distinguishes identical molecules by origin, because there is nothing to distinguish.

What does differ is everything that is not the peptide. Synthesis leaves its own impurity profile: chains that stopped short, chains missing a residue, fragments of protecting groups. Purification leaves a counter-ion behind. That is why a certificate measures the material rather than the molecule, and why two vials of the same compound are not automatically equivalent.

So what is a peptide made in a synthesiser? The same molecule, arriving with a different set of neighbours. Asking whether a peptide is natural or synthetic asks about the route; asking what is in the vial asks about the material, and only the second question has an answer on a document.

How peptides are made: solid-phase synthesis, one residue at a time

Almost every research peptide is built by solid-phase peptide synthesis, which assembles the chain while it is anchored to an insoluble resin bead. The first residue is attached by its C-terminus, and everything after that is a repeated cycle.

The cycle has three steps. A protecting group is removed from the amino end of the growing chain. The next amino acid, itself protected everywhere except the carboxyl group about to react, is coupled on. Then the excess reagent is washed away, which is the point of working on a solid support: the product stays on the bead and everything else goes down the drain. The chain grows from the C-terminus towards the N-terminus, the opposite direction to the one the finished sequence is read in.

An n-residue chain needs n minus 1 couplings once the first residue is loaded. Ipamorelin, at five residues, needs four. TB-500, at seven, needs six. BPC-157, at fifteen, needs fourteen. Tesamorelin, at forty-four, needs forty-three, each with its own deprotection and its own washes, so that synthesis is a few hundred separate operations before the chain is cleaved from the resin and purified.

Length is what makes a synthesis hard, and the reason is arithmetic. No coupling goes fully to completion. If each one reached 99%, a 15-residue chain with fourteen couplings would leave about 87% of the material as full-length product, and a 44-residue chain with forty-three couplings about 65%. Real syntheses use large reagent excesses and capping steps to push those figures up, but the direction is fixed: every residue added is another chance to fall short.

The failures have names that turn up in any purity discussion. A coupling that fails and is then capped leaves a deletion sequence, the full chain minus one residue. A chain that stops growing leaves a truncated fragment. Both are close structural relatives of the target, which is what makes them difficult to separate on a column and why peptide purity testing needs mass spectrometry alongside chromatography rather than chromatography alone.

After cleavage the crude material is purified by reversed-phase HPLC, and the solvent is then removed by freeze-drying, which is why the vial holds a lyophilized powder rather than a liquid or a crystal.

What a specification sheet tells you about a peptide

A specification is where the answer to what is a peptide turns into a short list of constants, each field answering a different question. None of them says anything about the batch in front of you, which is what a certificate is for, but together they pin down which molecule is described.

The sequence names the residues in order. The molecular formula counts every atom in one molecule: C62H98N16O22 for BPC-157, C38H68N10O14 for TB-500, C38H49N9O5 for ipamorelin, C221H366N72O67S for tesamorelin. That single sulfur in the tesamorelin formula is a sulfur-containing residue somewhere in the chain, the sort of detail a formula carries and a sequence makes you count for yourself.

The molecular weight is the mass of a mole of the peptide in grams, the same number as the mass of one molecule in daltons. It does the most work in practice, because it is what a mass spectrometer measures and what two similarly named compounds will differ on. It is also the number a peptide calculator needs if a concentration is to be expressed in molar terms rather than in milligrams.

The CAS number is a registry identifier assigned to one substance by the Chemical Abstracts Service, and it is useful precisely because trade names are not unique: 885340-08-9 refers to the acetylated seven-residue TB-500 fragment and to nothing else. A specification carrying a formula, a mass and a CAS number has said which molecule it means in three independent ways.

Then there is the field that explains why the powder in the vial weighs more than the peptide in it. Peptides are purified in acid and isolated as salts, most often trifluoroacetate, because reversed-phase purification runs in trifluoroacetic acid. That counter-ion is real mass sitting in the vial alongside the peptide, and so is residual water, since lyophilised material is hygroscopic. A 10 mg vial is 10 mg of fill, not 10 mg of peptide.

Purity and net peptide content are two different questions

This distinction catches most people. A vial contains peptide, counter-ion and water, and a certificate reports two numbers about it that are not two versions of the same measurement.

Purity is usually measured by reversed-phase HPLC and reported as area percent: of all the material the detector registered, what share was the target peptide? It is a ratio between peaks: how clean the peptide is relative to what else the detector saw.

Net peptide content is a mass. It answers how much of what is in the vial is peptide at all, and what it counts against the peptide is exactly what an area-percent figure cannot see: counter-ion, residual moisture, and anything else that does not absorb where the detector is looking. On the certificates behind this catalogue it is reported as a measured mass per vial against the label strength, which is more directly checkable than a percentage.

Neither number can be derived from the other, they do not average, and the larger does not stand in for the smaller. Across the 42 lots published on this site that report a content figure, the measured content runs from 93.6% to 131.6% of label with a median of 107.9%, because fills are set above the label strength so that every unit in a run clears its target once the filling tolerance is accounted for. A figure above the label is ordinary. A figure well below it, with no explanation beside it, is worth a question.

Reading one peptide certificate of analysis with those two fields in mind teaches this faster than any amount of definition does.

What is a peptide not? Where the word stops

A definition is sharper when it excludes things, and this market applies the word peptide to several things that are not peptides. The clearest example sits in this catalogue. NAD+ is not a peptide. It is nicotinamide adenine dinucleotide, a coenzyme built from two nucleotides joined by a phosphate bridge, with the formula C21H27N7O14P2. There are no amino acids in it and no peptide bonds. It is sold alongside peptides, supplied as a lyophilised powder in a vial like a peptide, and routinely described as one, which is why NAD+ peptide is a search term worth answering rather than correcting in silence. A molecule with no peptide bonds has a completely different chemistry from one that is nothing but peptide bonds.

Blends are a second case. A vial containing GHK-Cu, BPC-157 and TB-500 together is a mixture of three peptides, not a single peptide with a name of its own. It has no sequence, no formula and no molecular weight, because those fields describe one molecule and there are three in the vial. A blend certificate has to report the measured mass of each component separately.

Proteins are the third case, and there the word is being stretched. A 43-residue chain called a protein and a 44-residue chain called a peptide are the same kind of molecule, so the boundary is worth holding loosely. What is not loose is the difference between a seven-residue fragment and the full-length chain it was cut from: different molecules, different masses, whatever name they are sold under.

The test that settles all three cases is the same. Ask for the sequence, the formula and the molecular weight. A peptide has all three. A mixture has none of them. A coenzyme has a formula and a mass but no sequence, because it has no residues to list. That is the practical form of the question what is a peptide, and the products in this catalogue each publish all three fields alongside the certificate for the lot in the vial.