Plasma stability
A plasma stability assay incubates a compound in plasma at thirty-seven degrees and measures how much parent remains over time. For peptides it is one of the earliest informative experiments available, because proteolysis is usually the reason a potent peptide does nothing in vivo.
Running it so the number is interpretable
Use plasma from the species you intend to dose, and run human alongside it. Protease activity differs substantially between species, and rodent plasma is generally more aggressive than human. A peptide that survives four hours in human plasma and twelve minutes in mouse will look like a failure in the mouse study and may still be a viable human candidate.
Anticoagulant choice changes the result. EDTA chelates the divalent metals that metalloproteases require, so EDTA plasma systematically overestimates stability for any peptide cleaved by one. Heparin or citrate are the safer defaults, and the anticoagulant belongs in the method description because a comparison across two of them is not a comparison.
Use fresh or properly frozen plasma and keep the handling consistent. Protease activity declines with freeze-thaw cycles, so a peptide assayed against a much-thawed pool will look more stable than it is.
Work at a low concentration, one to ten micromolar. High concentrations saturate the proteases and flatter the compound.
Quench with cold acetonitrile to precipitate protein and stop the reaction at once. A quench that is slow or warm keeps digesting while you work through the plate.
Controls that decide whether you believe it
Run a positive control peptide with a known short half-life on every plate. Plasma lots differ, and without it you cannot tell a stable compound from a dead plasma lot.
Run a buffer-only arm alongside. Loss in buffer is chemical degradation or adsorption rather than proteolysis, and the two need different fixes. A compound that disappears equally in both was never a protease problem.
Measure the parent by LC-MS on the specific mass rather than by total UV area. Fragments absorb at 214 as well, so a UV-only readout can show a flat trace while most of the parent is gone.
Reading the fragments
The most useful output is often not the half-life. Identifying the cleavage sites tells you where to intervene, and it costs one extra injection with a longer mass range.
Exopeptidases account for a large share of peptide turnover in plasma. Aminopeptidases trim from the N terminus and carboxypeptidases from the C terminus, and both are blocked cheaply: N-terminal acetylation, C-terminal amidation, or a D-amino acid or unnatural residue at the terminal position.
Endopeptidase cleavage shows up as two internal fragments. That needs a targeted change at the scissile bond, an N-methylation, a D residue, or a backbone constraint such as a lactam bridge or macrocyclization.
Why the in vitro number underpredicts
Plasma stability measures only proteolysis in blood. It omits the two processes that usually dominate real exposure.
Renal clearance removes anything below roughly five kilodaltons by glomerular filtration, regardless of how protease-resistant it is. A peptide can be perfectly stable in plasma and still have a circulating half-life of minutes.
Tissue and membrane-bound peptidases, particularly in liver, kidney and the vascular endothelium, are absent from a plasma tube and contribute substantially in vivo.
Plasma protein binding runs the other way and extends exposure, by shielding the peptide and reducing the free fraction available to both proteases and filtration.
The practical use of the assay is therefore comparative. It ranks analogs quickly, it identifies the cleavage sites worth engineering around, and it catches the compounds that are hopeless. Treat the half-life as a screening number and let the pharmacokinetic study set expectations for exposure.