Every amino acid a ribosome uses comes in one handedness, and every protease evolved to match it. That single fact is the reason D-amino acids appear in so many engineered peptides: put the mirror-image residue at the position an enzyme would cut, and the enzyme no longer recognises the site. It is the cheapest stability modification there is — one residue swapped for its reflection — and it appears in Melanotan-2, every growth hormone releasing peptide, and a good part of the rest of the catalogue.
What handedness means
Nineteen of the twenty standard amino acids have a carbon with four different groups attached, which means they exist as two non-superimposable mirror images. By convention they are labelled L and D. Biology uses L almost exclusively: the ribosome builds L chains, and the enzymes that cut peptides have active sites shaped for L backbones.
Glycine is the exception. Its side chain is a hydrogen, so its central carbon carries two identical groups and it has no handedness at all. There is no D-glycine, which is why it never appears in these discussions.
| Property | L residue | D residue in the same position |
|---|---|---|
| Mass | Identical | Identical |
| Chemical composition | Identical | Identical |
| Recognised by proteases | Yes | Poorly or not at all |
| Backbone geometry | Native | Locally reversed; may favour a turn |
| Receptor binding | Native | Changed; has to be measured |
| Distinguishable by mass spectrometry | No | |
Why does a D residue block a protease?
Because the active site is a three-dimensional pocket shaped to hold an L backbone in a specific orientation, with the side chain pointing one way and the carbonyl positioned for attack. A D residue at that position points its side chain the other way and puts the carbonyl in the wrong place. The substrate no longer fits, and the bond is not cut. A single D residue protects one site; protecting a peptide with several cleavage sites needs several substitutions or a different strategy. It is one of four routes, and it is the most surgical.
Does the substitution change what the peptide does?
Usually, and not always in the direction intended. A D residue reverses the local backbone geometry and often favours a turn, which changes the shape presented to a receptor. Sometimes that is neutral. Sometimes it improves binding, because the turn happens to be the active conformation — the D-phenylalanine in the growth hormone releasing peptides is thought to work this way. And sometimes it abolishes activity, because the new shape does not fit. Every D substitution is therefore two experiments: does the peptide survive, and does it still bind. The literature reports the ones that passed both.
Where the notation gets confusing
A D residue is written with a prefix — D-Phe, D-Trp — or, in single-letter code, as a lower-case letter: f for D-phenylalanine against F for the L form. Some sources omit the distinction entirely, which makes a sequence ambiguous in a way that matters. Melanotan-2 written without its D-Phe marked describes a different molecule. Reading the notation around a sequence includes reading case.
Can a certificate tell L from D?
Not by mass, and this is the important gap. L and D residues weigh the same, so the identity panel confirms the composition and says nothing about handedness. Neither does a standard reversed-phase purity trace, unless the two forms happen to separate — which they sometimes do as a small shoulder, since a diastereomer is a slightly different shape. Establishing stereochemistry requires a chiral analysis: hydrolysing the peptide and running the free amino acids on a chiral column, or a chiral derivatisation. Most certificates do not include it. For a peptide whose activity depends on a D residue, the absence of that test means the handedness is assumed from the synthesis rather than confirmed on the batch.
Where does an unwanted D residue come from?
From the synthesis. Each coupling step activates a carboxyl group, and an activated residue can lose its stereochemistry briefly — racemise — before the bond forms. A small fraction of chains then carry one residue in the wrong handedness. Synthesis runs C to N partly to keep this rare, and certain residues (histidine and cysteine especially) racemise more readily than others. The result is a diastereomer impurity: same mass, same composition, nearly the same retention time, different shape. It sits just beside the main peak on a chromatogram, and a shoulder there is the place to suspect it.
What is a retro-inverso peptide?
The extreme case of the same idea. Reverse the sequence order and make every residue D, and the side chains end up pointing in approximately the same directions as in the original — the backbone runs the other way but the surface presented looks similar. The result is a molecule proteases cannot touch at all, since every residue is the wrong hand, that sometimes retains binding because the side-chain arrangement is preserved. It works for some sequences and fails for others, for the same reason a single D substitution does: the backbone geometry is not identical, only approximately so, and a receptor decides how close is close enough.
Are D-amino acids found in nature at all?
Yes, in small amounts and specific places. Bacterial cell walls contain D-alanine and D-glutamate, which is part of why they resist the host's proteases. Certain frog skin peptides and cone snail venoms carry a single D residue installed by an enzyme after the chain is made. And D-aspartate accumulates slowly in long-lived human proteins through spontaneous racemisation over decades. None of this changes the working rule: a ribosome makes L chains, and a D residue in a synthetic peptide was put there on purpose.
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.
