In February 1935, inside the DuPont Experimental Station in Wilmington, Delaware, a chemist named Gerard Berchet made a polymer that could be drawn into a fiber thinner than a hair and stronger than silk. His supervisor, Wallace Carothers, had been hired to run basic research with no product required. The result became nylon, and the molecule behind it shares a defining feature with every peptide ever made: a repeating amide bond.
That shared chemistry is the shortcut to understanding something that confuses a lot of readers. The words peptide, protein and drug sound like separate worlds. In practice they sit on one spectrum, and the place where a chain stops being one thing and starts being another is set by length, a number that regulators have written into law.
A polymer built from the same bond as a peptide
DuPont hired Carothers in 1928 to lead fundamental polymer research at its Wilmington laboratories. Archival descriptions of his work at the Hagley Museum and Library note that he was trying to prepare polymers structurally analogous to natural materials such as cellulose and silk. Silk is a protein, which makes it a long chain of amino acids joined by amide links.
Nylon is a polyamide. Its backbone repeats the same linkage, formed when an amine group meets an acid group and a molecule of water is released. A peptide bond forms the same way. The difference lies in the building blocks. Biology works from twenty amino acids, each with its own side chain, and nylon is assembled from two simple industrial chemicals. One is a diamine, the other a diacid.
DuPont announced nylon publicly in 1938. By then the Wilmington group had shown that long molecules built from repeating units behave according to rules that can be studied and predicted, and that principle carries straight through to peptide science. Link enough of the same bond together and the properties depend as much on length as on chemistry.
Water cuts what water joined
The reaction that builds an amide bond can run backward. In the presence of water, with the right conditions, the bond is cleaved. This is hydrolysis, and it matters in three places at once: in the body’s own digestion of protein, in industrial processing, and in the storage of dried material that was never meant to meet moisture.
Collagen shows the practical consequence. Whole collagen is a very large protein. A collagen hydrolysate is what you get when that protein has been cut into fragments, and the fragments are not uniform. A material sold as collagen peptides is a population of pieces of many lengths, so a single average on a specification sheet only summarizes it. Peptides Wilmington’s page on how chain length changes what a collagen peptide does starts from the same observation: the chain length of a collagen peptide changes how it is digested and absorbed, which is why two products from the same source can behave differently.
The lesson generalizes. When a label says peptide, a good follow-up question is how long, and how uniform.
Where the United States draws the line
For most of the twentieth century, a protein product could be approved as a drug or as a biologic depending on its history. In 2020 the rules changed. A federal statute amended the definition of a biological product by removing an exception for chemically synthesized polypeptides, and the FDA responded with a final rule published in February 2020. It interprets the word protein to mean any alpha amino acid polymer with a specific, defined sequence that is greater than 40 amino acids in size.
The effect was immediate. On March 23, 2020, products that had been approved under new drug applications and met that definition were deemed to be licensed as biologics. Insulin was the best-known example. The FDA counted its two chains together, 51 amino acids in all, which put it comfortably over the line. Human growth hormone was another.
Chains at or below 40 amino acids fall on the other side. Many of the metabolic peptides in the news sit just under it, which is part of why they are approved and regulated as drugs rather than biologics. A short explainer on the regulatory status of metabolic peptides opens from the broader point that these are prescription medicines in the United States, and the length rule is one reason the paperwork around them looks the way it does.
A bright line is not a law of nature
The 40-amino-acid figure deserves a careful reading. In its rulemaking the FDA acknowledged that some commenters pointed out that proteins can be smaller than that, and it kept the number anyway. The agency described its choice as a clear, workable cutoff that avoids arguments over folding and function. The chemistry does not change at amino acid 41. A regulator needed a rule that a manufacturer could apply on day one, and length was the rule that could be counted.
For readers, the useful point is that categories in this field are conventions layered on a continuum. A 39-chain compound and a 41-chain compound may resemble each other closely in a laboratory, and still sit under different legal frameworks with different approval pathways, different manufacturing expectations and different language on the label.
What a 1930s polymer chemist would ask today
Carothers was known for pairing experiment with theory. The American Chemical Society’s history of his work notes that he believed mathematics could be applied to understand how polymers form and what properties result. That habit translates into a few plain questions for anyone reading a peptide document.
- Is the sequence stated, and is its length given as a number?
- Does the report describe a single defined compound, or a distribution of fragment sizes?
- Which method measured the result, and which laboratory ran it?
- Does the claim rest on a study in people, or on an experiment in cells?
The last question catches more errors than the first three combined. Popular writing about skin and hair peptides routinely treats a result in cultured cells as if it were a result in people. Peptides Wilmington’s copper peptide reading list is built around separating those two kinds of evidence, and the method it models is the right one: find the primary source, read what was measured and in whom, and only then decide what the headline is worth.
What this article does not claim
It does not say that any particular product is safe, effective or lawful for use by people. It does not describe how any compound is made at scale, stored or handled. Its subject is vocabulary: why chain length, a number anyone can read on a specification, ties together an industrial polymer, a collagen supplement and a hormone that millions of patients rely on.
The materials discussed here are laboratory reagents or investigational substances where they are not approved medicines, and nothing in this piece is guidance for use on people.
The number worth looking for
The Wilmington chemists proved that long molecules obey rules, and the rules start with how many units are in the chain. Regulators later wrote that logic into law with a figure of 40. A reader who learns to look for chain length, and to ask how it was measured, can read a label, a study and a regulatory notice with the same set of eyes.