Natural, Synthetic, and the Marketing In Between

On the screen of a chromatography workstation, two traces have been laid one on top of the other. One comes from a peptide expressed by engineered yeast in a fermentation tank and purified from the broth. The other comes from the same sequence assembled one amino acid at a time on plastic beads in a synthesiser. The main peaks sit at the same retention time. The mass spectrometer gives the same molecular weight for both, to the decimal place. Asked which one is natural, the analyst shrugs, because the instrument cannot answer that question and neither, in any useful sense, can the analyst.

So are peptides natural or synthetic? The question assumes the words describe the molecule, and they do not. They describe the route it took. The same chain of amino acids can be extracted from animal tissue, expressed in engineered cells or built by chemical synthesis, and if each route is carried out correctly the products are the same molecule. Where the routes genuinely differ is in what else comes along: the by-products and residues each process leaves behind. Regulators care a great deal about that, and almost not at all about whether a product's origin was natural. Marketing is the reverse, and this piece is about the gap between the two.

Three things "natural" can mean

When the word appears near a peptide, it is usually doing one of three quite different jobs, and it helps to separate them.

The first meaning concerns the sequence: the chain occurs in the human body. Many research peptides do resemble something endogenous, meaning made inside the organism. Some are exact copies of hormones or signalling fragments. Many more are analogues, deliberately altered with a swapped residue, a mirror-image amino acid, a capped end or an attached fatty acid chain, precisely so that they behave differently from the natural version. An analogue designed to resist the enzymes that destroy the natural molecule is, by construction, not the natural molecule. Some catalogue peptides are described as fragments of naturally occurring proteins even though nobody has shown that the fragment itself circulates in anyone.

The second meaning concerns the source: the material came out of a living thing, as insulin once came from the pancreases of slaughtered animals. The third meaning concerns nothing at all. It is the word as an aura, a leaf printed on a pack, doing emotional work with no chemical content. The first two meanings can be checked. The third cannot, which is why it is the one marketing prefers.

The molecule does not remember where it came from

A peptide's identity is its sequence plus any modifications to it: which residues, in which order, with which chemical groups attached and in which three-dimensional arrangement. Two samples that match on all of those are the same substance, however they were produced. There is no residual trace of the fermentation tank in a correctly purified recombinant peptide, and nothing mechanical about a chemically assembled one. Atoms do not carry their history.

What differs between routes is the company the molecule keeps. Chemical synthesis builds a chain in repeated cycles, and each cycle that falls short leaves behind shorter chains with a residue missing, as well as fragments still carrying remnants of the protecting groups used to control the chemistry, and counter-ions picked up during purification. Expression in cells leaves the opposite kind of residue: proteins and DNA from the host organism and, with bacterial hosts, endotoxin from their cell walls. Extraction from tissue brings along whatever else the tissue contained. Each route has its own characteristic impurity profile, and that profile, not the origin itself, is the meaningful difference.

This is the logic of the European Medicines Agency's guideline on the development and manufacture of synthetic peptides, adopted at the end of 2025 and in force from June 2026. It is a document about pharmaceutical manufacturing quality, not about research chemicals, and it does not rank synthetic material as better or worse than any other kind. It asks manufacturers to understand and control the impurities their particular process creates 3. The question it puts to a manufacturer is not where the molecule came from but what else is in the product.

When natural was the dangerous option

History offers two cases that should unsettle anyone who treats natural as a synonym for safe. The first is insulin. For its first sixty years, the insulin that kept people with diabetes alive was extracted from the pancreases of cattle and pigs. Pig insulin differs from the human hormone by a single amino acid, and cattle insulin by three. In 1979 a team reported that they had made synthetic genes for the two chains of human insulin and got bacteria to express them 5. Within a few years the product of that approach was on the market. The insulin made in bacteria was identical in sequence to the one made by a human pancreas, and the natural product was the one that was slightly foreign.

The second case is darker. From the late 1950s until 1985, children with growth disorders in several countries were treated with human growth hormone extracted from pituitary glands collected at post-mortem. Pooled glands from very many donors went into each batch. Some of those glands carried the prion agent that causes Creutzfeldt-Jakob disease, and over the following decades more than two hundred recipients worldwide developed it, often many years after treatment 6. Extraction ended when a recombinant version became available, a product that by its nature could not carry a human prion. In this case the natural product was the one that killed people.

Regulators classify by what, not where from

Medicines law reached the same conclusion by a longer route. In the United States, whether a product counts as an ordinary drug or a biological product used to depend partly on how it was made: the statute defining biological products excluded chemically synthesised polypeptides. In 2020 Congress removed that carve-out, and the FDA's accompanying rule defined a protein by length alone, as a chain of more than forty amino acids with a specific, defined sequence 2. Since then, whether a chain was grown in cells or assembled by chemistry has no bearing on which regulatory regime it falls into. Its length does.

The regulator's other questions are just as indifferent to origin. What is the substance, precisely? How is it made, and is the process controlled? What is claimed for it? A peptide presented as a treatment is regulated as a medicine whether it came from a pig, a yeast or a resin. A peptide supplied for laboratory research is a reagent on the same terms. At no point in any of these frameworks does a product earn a lighter touch by being natural.

Why the word works anyway

If the distinction is chemically empty and legally irrelevant, why does it sell? Psychologists have measured the preference directly. In a 2004 study with university students and a representative sample of American adults, a large majority preferred natural versions of foods and medicines. The striking finding was what happened when the researchers took away every practical reason for the preference. When participants were told that the natural and artificial versions were equally healthy and effective, or even chemically identical, most of those who preferred natural went on preferring it 1.

The authors concluded that much of the preference is ideational rather than instrumental: it rests on moral or aesthetic ideas about what natural means, not on beliefs about outcomes 1. Interestingly, the preference was weaker for medicines than for foods, as though people sense that pharmacology is a domain where the natural and the effective part company. For marketing, that weaker preference is a problem, and the solution is to dress the medicine up as something closer to food: a supplement, a wellness product, a molecule your body already makes.

A word with a history: bioidentical

The most sophisticated version of this move has a name. "Bioidentical" arose in the market for compounded hormone therapy, where it was used to describe hormones with the same structure as those the body makes, and it carried a strong implication that such products were therefore safer or more natural than approved ones. The Endocrine Society, reviewing the field in 2016, was blunt about the gap. Many approved, regulated hormone products are themselves structurally identical to human hormones, so the word distinguishes nothing. Its real function had been to suggest a safety advantage for compounded preparations that evidence did not support, while those preparations escaped the testing, labelling and batch control required of approved products 4.

The word has since migrated to peptides, with the same structure of implication. Bioidentical asserts something about a sequence and hopes the listener will infer something about safety, purity and regulation. The inference does not follow. Structural identity says nothing about how a batch was made, what came with it, or whether anyone has tested it. The regulated insulin in a pharmacy fridge is exactly as bioidentical as anything on a wellness website, and it comes with a batch record.

The better question

Strip away the marketing and one serviceable question remains, and it has nothing to do with nature. It is not "where did this come from?" but "what else is in it?" For a synthetic peptide that means the truncated and deleted sequences, the residual reagents and the counter-ion. For a recombinant one it means host-cell proteins, DNA and endotoxin. For anything extracted it means whatever the source tissue held. Every one of those questions has a measurable answer, with a method, a limit and a result on a certificate.

That is where a piece about language has to stop, and where analytical chemistry begins: the chromatogram, the mass spectrum and the impurity table that tell you what a particular batch actually contains. Those two overlaid traces on the analyst's screen were never going to say which molecule was natural. What they could say, if anyone looked at the small peaks around the big one, is which product was clean. That was always the more useful thing to know.