Potent, but not precise

The body never cut a key for one door. It wrote a single word — protect, feed, calm — and trusted each tissue to listen for its own. Our drugs are still learning that economy.

Ask why there is no clean tanning drug — a peptide that safely, precisely turns the skin’s pigment switch and nothing else — and the honest answer begins by correcting the question. The potent part is solved. What is missing is precision, and precision turns out to be the harder thing to buy.

Potent is solved. Selective is not.

There is already an approved drug that switches on the pigment receptor MC1R and reliably darkens skin: afamelanotide, a copy of the body’s own α-MSH re-engineered to be far more potent and far longer-lasting than the natural hormone. In fair-skinned volunteers it measurably increased pigmentation and cut UV damage.^[5] The grey-market tanning peptide Melanotan II is another potent one, a cyclic analogue designed in the 1980s to be more stable and more active than α-MSH;^[3] the investigational pill dersimelagon is a third. Potency — making the receptor fire hard — was never the obstacle.

The obstacle is that not one of these hits MC1R alone. Every one of them also switches on the other melanocortin receptors, and that is why the tan never comes by itself. To see why, you have to look at what these molecules are actually recognising.

One key, five locks

Every melanocortin — α-MSH, ACTH, the drugs built to imitate them — carries the same tiny recognition core: four residues, His-Phe-Arg-Trp. That short sequence is the working end of the signal, and all five receptors evolved to read it.^[1] A small peptide built around that core is, in effect, a key cut to a shape that four of the five locks accept. α-MSH itself is completely non-selective — it activates MC1R, MC3R, MC4R and MC5R at broadly similar strengths.^[1][2]

The lone exception proves the rule. MC2R, the adrenal receptor, does not answer to that four-residue key at all — it reads only the longer ACTH, and it cannot even reach the cell surface without a chaperone protein, MRAP1.^[10] The one receptor that escaped the family resemblance did so by needing an entirely different key and a helper to use it. The other four are close enough that a single molecule reads them all.

The body never needed a selective one

This looks like an oversight until you notice how the system actually achieves precision. Evolution never made a MC1R-only ligand because it never had to: specificity comes from where a peptide is released and which receptor is switched on in that tissue, not from the molecule telling the receptors apart.^[1] MC1R sits on pigment cells and immune cells; MC4R sits in the appetite circuits of the brain; MC2R sits on the adrenal gland. The same broad signal means very different things in each place, and the body keeps them straight by geography. There was no selection pressure for a molecularly selective key — so there is no natural template for one to copy.

Potency and selectivity are different knobs

Here is the part that defeats the intuition that a “better” drug would be a cleaner one. The classic tricks that make an α-MSH copy powerful — the substitutions behind NDP-α-MSH and the cyclic Melanotan analogues — raise activity at all the receptors at once, because they strengthen the grip on the shared core, not on any one pocket.^[4][2] So afamelanotide is potent and broad; Melanotan II is potent and broad. Turning the dial up does not turn precision on. You can watch this spill the other way too: setmelanotide, engineered for the appetite receptor MC4R, still grips the pigment receptor MC1R almost as tightly, so skin darkening is one of its most common effects.^[7] The binding matrix lays out those receptor-by-receptor numbers.

Selectivity can be coaxed out of these molecules — but the margins are exquisite. Take the potent tanning peptide, change a single residue at position seven to a bulky unnatural one, and it stops activating MC3R and MC4R and startsblocking them — while still switching MC1R on. That is exactly what the research compound SHU9119 does.^[8] One atom’s worth of change flips the molecule from agonist to antagonist at some receptors and not others. It proves subtype discrimination is chemically possible; it also shows how narrow the ledge is, and how easily a change aimed at one receptor lands on another.

A selective one exists — for the other reason

In fact the clean molecule already exists. BMS-470539 is a small-molecule agonist that switches on MC1R and largely spares the rest of the family.^[9] But read what it was built for: it was developed to test whether firing MC1R could calm inflammation — it was dosed to block cytokine release and stop white cells flooding inflamed tissue in mice, and it lives in the literature as an anti-inflammatory tool. Nobody took it toward pigmentation. The selectivity that a “clean tanning drug” would need is not a chemistry that can’t be done. It is a chemistry that was pointed somewhere else, because that is where the medicine was.

There is no medical reason to build it

Follow the incentive and the gap explains itself. MC1R agonism for pigment has exactly one approved use — erythropoietic protoporphyria, a rare disease in which sunlight causes agonising pain — and afamelanotide already serves it, delivered as an implant where broad activity is tolerated because the photoprotection is worth it.^[6] The enormous market is cosmetic tanning, and that is not a road drugs are approved down; the grey-market history of Melanotan II — contaminated vials, case reports of changing moles — is a large part of why no serious developer wants to plant a flag there.

Where the hard medicinal chemistry of melanocortin selectivity is being done, MC1R is usually the receptor to avoid. The money is in MC4R for obesity, and for those drugs the tan is an unwanted side effect — so the newest agents are engineered to spare MC1R, not to hit it. The field is spending its selectivity budget teaching drugs to miss the pigment receptor. The full landscape of who is building what sits on the melanocortin agonists page.

Even a perfect one would meet a ceiling

Suppose someone did build the flawless MC1R-selective peptide. Two ceilings would still be waiting. The first is delivery: peptides are digested if swallowed and clear quickly if injected, which is why every approved melanocortin peptide is an injection or an implant, and why the frontier has moved toward oral small molecules in the first place. The second is biology. MC1R agonism only produces pigment if the receptor can answer: the common red-hair variants of MC1R are loss-of-function, sitting at reduced levels on the cell surface and signalling weakly,^[11] and the people who most want a tan are disproportionately the people carrying them. You cannot out-dose a receptor that cannot fire, and past a point the melanocyte’s own machinery caps how much eumelanin it will make.^[1] A more potent key does not linearly make more colour.

Why this is finally changing

For most of this story the tools did not exist to design a selective melanocortin on purpose. That changed in 2020–21, when cryo-electron microscopy resolved the receptors at atomic detail — including MC1R with α-MSH and afamelanotide in its pocket — and made the near-identical pockets, and the small differences between them, finally visible.^[10] Structure-guided redesign has already turned that into selectivity: reworking setmelanotide’s core produced a compound with more than a hundred-fold preference for MC4R over MC1R — a deliberate, successful attempt to separate two receptors that used to come as a pair.^[12] The full structural story is on the receptor structures page.

That is the honest state of it. A selective MC1R agonist is no longer a chemistry problem the way it was a decade ago — the pictures are in hand and the design methods work. What is missing is not the key. It is a reason to cut it: an indication, a market, a medicine worth making. Until one appears, “potent” will keep arriving without “precise,” and the tan will keep coming with everything else attached.