CAS Numbers For Peptides, And Why Salt Form Changes Them

The format is fixed: two to seven digits, a hyphen, two digits, a hyphen, and a single check digit. Nothing in the first two blocks means anything. Numbers are issued in the order substances were registered, so a low number indicates an early registration and a high one a recent registration, and that is the full extent of the information in the digits themselves.

This matters because it rules out one thing people reasonably expect of an identifier. A registry number cannot be checked against a structure, a formula or a molecular weight, because it is not built from any of them. Comparing a number to a substance means consulting the registry. There is no computation that recovers one from the other.

For peptides this creates a specific difficulty. The registry holds separate entries for the free base, for each salt, and for each hydrate of the same sequence, because each of those is a distinct substance with a distinct molecular formula. A sequence commonly supplied as an acetate has at least two registry numbers in circulation, and material described only by sequence has neither of them attached to it unambiguously.

The check digit, and how to test one

The final digit after the second hyphen is a modulo-10 checksum. The procedure: drop the check digit, read the remaining digits from right to left, multiply the first by 1, the second by 2, the third by 3 and so on, sum the products, and take the remainder on division by 10. That remainder must equal the check digit.

Worked on water, registry number 7732-18-5. Dropping the check digit leaves 773218. Read right to left: 8, 1, 2, 3, 7, 7. The products are 8 × 1 = 8, 1 × 2 = 2, 2 × 3 = 6, 3 × 4 = 12, 7 × 5 = 35 and 7 × 6 = 42, summing to 105. The remainder of 105 on division by 10 is 5, which matches. The number is well formed.

The test is worth running on any registry number copied from a document, because it catches the two most common transcription errors. A single wrong digit changes the weighted sum by a multiple that is almost never a multiple of 10, and a transposed adjacent pair changes it by the difference between the two digits. Both usually break the checksum. What the test cannot do is tell anyone whether a well-formed number points at the intended substance, because a different substance's number is also well formed.

Why the salt gets its own number

CAS registers substances as reported, and a salt differs from its free base in molecular formula. Acetate, trifluoroacetate and hydrochloride salts of one peptide are three substances, and a hydrate of any of them is a fourth. The registry reflects that, so a single sequence can be associated with several numbers, none of which is more correct than the others in the abstract. Which one applies depends entirely on what is in the container.

Synthesis history is what decides it. Peptides purified by reverse-phase chromatography with trifluoroacetic acid in the mobile phase leave that separation as trifluoroacetate salts, and converting to acetate or hydrochloride requires a deliberate exchange step. The counterion in the vial therefore reflects the last ion-exchange or lyophilisation step, not the sequence.

The failure mode: free-base mass used on salt material

The failure to name is the one that follows directly from all of the above. A registry number is looked up, the molecular weight listed against it is taken as the molecular weight of the material, and a labelled mass is converted into moles using it. If the number was the free base and the vial holds a salt, the conversion is wrong by the mass of the counterion, and it is wrong in the direction that overstates how much peptide is present.

How it shows up: two quantities derived from the same vial disagree by a few percent and neither can be made to reconcile. Because the error is systematic rather than random, repeating the calculation does not reveal it, and because a few percent sits inside the spread of most casual measurements, it often goes unnoticed for a long time.

The molar ratio is not a free choice either. It tracks the count of basic side chains available to carry a positive charge, so a sequence rich in arginine and lysine supports more counterion than one with few basic residues. Counterion content is measured, where it is measured at all, by ion chromatography or by fluorine-19 nuclear magnetic resonance against a fluorinated internal standard, and reported as percent by mass.

What a registry number does and does not certify

• It names a substance in an index. That is its whole function, and it does it well.
• It does not encode purity. Two lots of the same substance at very different purity share one number.
• It does not encode stereochemistry beyond whatever CAS registered, so a racemised position may or may not be reflected by a separate entry.
• It does not encode a terminal modification unless that modification is part of the registered substance, which is why a written sequence with its terminus annotated carries information a number alone does not.
• It is not a traceability record. A lot number identifies material; a registry number identifies a substance class.

The last distinction is the one most often collapsed. Substance identity and lot identity answer different questions, and a specification sheet needs both because neither substitutes for the other. The same goes for the terminus: a written sequence with its terminus annotated says something the registry entry may not.

Where our own position stops

Specific registry numbers for catalogue items are not reproduced in this journal, and the reason is the same one that governs how we handle citations: a transcribed identifier that turns out to point at a different substance is worse than no identifier at all. A registry number belongs in the registry, where it can be checked, rather than restated from memory in prose.

Two further limits are worth stating plainly. Counterion content is not among the specifications Aurum publishes, so we cannot state which salt form a lot is or what fraction of its mass is counterion, and no purity figure of ours carries that information. And our published purity is chromatographic area percent, not the fraction of a vial's mass that is peptide, which is precisely the quantity a salt-form question is really asking about.

Common questions

Can a CAS number be worked out from a structure?

No. Numbers are assigned sequentially at registration and carry no structural information, so the only route from structure to number is a registry lookup.

Why does a search return several numbers for one peptide?

Because the free base, each salt and each hydrate are separate registered substances. All of the numbers are real; they describe different material.

Does a number ever get retired or merged?

Entries can be deleted or superseded when a registration is found to duplicate another or to have been based on a mistaken structure, so an old number copied from an old document may no longer be the current entry.

Is a CAS number a substitute for a sequence?

No. A sequence written out in one-letter or three-letter code with both termini annotated is a complete structural statement; a number is a pointer that requires the registry to resolve.

Would a D-amino acid at one position change the number?

A peptide carrying a D-amino acid at one position is a different substance, so in principle yes. Whether a separate entry exists depends on whether that stereoisomer has been registered, which is a reason to state stereochemistry explicitly rather than rely on the number.