Most of what is known about research peptides was learned in rodents, and rodents do not make the same peptides we do. Some hormones are identical across species; others differ by a residue or two in exactly the region a receptor reads. A result from a mouse is a result about a mouse receptor meeting, in many cases, a human peptide sequence — or the reverse. Knowing which pairs match and which do not is the difference between reading a rodent paper and misreading it.
Where the sequences agree and where they do not
| Peptide | Human vs rodent sequence | Practical consequence |
|---|---|---|
| GLP-1 | Identical | Human analogues act on rodent receptors as expected |
| GIP | Differs at two positions | Potency at the rodent GIP receptor is not the human figure |
| Amylin | Differs at several positions; rodent amylin does not form amyloid | Aggregation findings do not transfer either way |
| GHRH | Substantially different | Human GHRH is a weak agonist at some rodent receptors |
| Ghrelin | Differs at two positions | Minor; the acyl group matters more |
| Kisspeptin-10 | Differs at one position | Both active at both receptors; potency differs |
| α-MSH | Identical | Melanocortin analogues transfer well at the ligand level |
The pattern is that short, ancient signalling peptides tend to be conserved and longer hormones tend to drift. That is a tendency, not a rule, and the table is the thing to check rather than the tendency.
Why does a two-residue difference matter?
Because the residues that differ are often the ones the receptor contacts. A hormone and its receptor co-evolve, so a species-specific change in the hormone is usually matched by a change in the receptor's binding pocket. The human hormone then fits the rodent receptor slightly wrongly, and the rodent hormone fits the human receptor slightly wrongly, and "slightly" can be an order of magnitude in affinity. GIP is the standing example: tirzepatide's GIP potency reported in a rodent system describes a different receptor from the one it was designed for.
Is the receptor the bigger variable, or the peptide?
Both, and it is worth separating them. A synthetic human-sequence peptide applied to a rodent is a human ligand on a rodent receptor. A rodent given its own hormone is a rodent ligand on a rodent receptor. Neither is a human ligand on a human receptor, which is what an in-vitro assay on a human cell line provides. The three systems can rank the same set of compounds differently, and a paper's system is the first thing to identify before its numbers are compared to anything.
Beyond the sequence
Even where the peptide and receptor match, the system around them does not.
Clearance scales with body size. A peptide's half-life in a mouse is routinely several times shorter than in a human, because small animals clear everything faster. A compound that needs modification to persist in a human may need none in a rodent, and vice versa.
Enzymes differ. Rodent DPP-4, neprilysin and the serum peptidases have their own specificities, so a native peptide's fate in rodent serum is not its fate in human serum. The degradation table is species-specific in its details.
Receptor distribution differs. Where a receptor is expressed, and how densely, varies between species. A response that runs through a tissue rich in the receptor in mice may be weak in a species where that tissue expresses little of it.
Does this mean rodent data is unreliable?
No — it means rodent data is data about rodents, which is what it was always claimed to be. The literature on almost every compound in this catalogue is built on it, and it is where mechanisms are found and hypotheses formed. The error is not in the data but in reading a rodent result as if it described a human system, without checking whether the peptide, the receptor, the enzymes and the distribution match. Often several of them do not, and the honest reading is that the result establishes a mechanism in one system and raises a question about the other. Compounds whose evidence is almost entirely rodent carry this caveat on every claim.
What about the peptides with no natural counterpart?
Fragments and designed sequences raise a different version of the same question. Dihexa or AOD-9604 have no rodent or human native form to compare; what differs across species is only the receptor and the system. That is simpler in one way — one variable rather than two — and harder in another, because there is no native hormone whose known cross-species behaviour can serve as a guide. For these, the receptor's conservation is the thing to look up.
How should a rodent result be cited in a research write-up?
With the species named, the sequence used identified as human or rodent, and the receptor system stated — and with the comparison to any human-system result made explicitly rather than by implication. A sentence such as "human-sequence peptide X reduced Y in mice" is a complete and honest description; "peptide X reduces Y" is the same sentence with the information removed. The system a result came from is part of the result, and for cross-species work it is the part most often dropped.
Is there a quick way to check whether a sequence is conserved?
Yes: compare the human and rodent sequences directly, which any protein database provides in a minute. Where they are identical, the ligand variable disappears and only the receptor and the system remain. Where they differ, note which positions, and then look at whether those positions sit in the region known to contact the receptor — a difference at the C-terminus of a peptide whose receptor reads the N-terminus may not matter. It is a five-minute check that, done before reading a rodent paper, changes how much of it can be carried across.
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.
