Peptide Stability in Serum and Plasma: How In Vitro Assays Work

What the assay asks

Blood and its fractions contain many proteases, enzymes that catalyse the hydrolysis of peptide bonds. A serum or plasma stability assay asks a narrow question: under defined conditions, how fast does a given peptide disappear, and what fragments appear? Slower chemical reactions such as deamidation and oxidation are covered in the storage stability guide.

Serum and plasma stability has long been used as a screening assay to eliminate unstable candidates. For most L-amino acid peptides in one such series, the main route of breakdown was cleavage by exopeptidases, which remove residues from the ends of the chain.^[1] The spread of results is wide. In a comparison of several peptides in diluted human serum and other biological media, half-lives ranged from about 4 minutes to more than 10,000 minutes, with most between about 10 and 100 minutes.^[2]

Serum, plasma and whole blood are not the same matrix

Serum is the supernatant left after blood has fully clotted. Plasma is the supernatant from blood that was mixed with an anticoagulant and centrifuged. The difference matters for proteolysis: clotting activates proteases and cofactors, while anticoagulants inhibit some enzymes, EDTA by removing metal ions that metallo- and calcium-dependent proteases need, and heparin by blocking thrombin and factor Xa.^[3]

One study collected mouse blood directly into the peptide solution, so that clotting was avoided and degradation could be followed in fresh whole blood for an hour, and compared peptides from three families with fresh serum and commercial serum and plasma. In general, peptides were degraded faster in serum than in plasma, all were more stable in fresh blood, and the order of degradation rates among the peptides changed across the six incubation conditions. The authors cautioned that stabilising a peptide at a site cut only in serum or plasma assays may be of limited relevance.^[3]

A worked example: GLP-1 and DPP-4

The degradation of glucagon-like peptide-1 shows what an assay can reveal. When GLP-1(7-36) amide was incubated in human plasma at 37 °C, an N-terminally truncated peptide formed with a half-life of about 20 minutes. The reaction was unaffected by EDTA or aprotinin but was completely prevented by inhibitors of dipeptidyl peptidase-4 (DPP-4) or by keeping the plasma at 4 °C. Mass spectrometry and sequence analysis identified the product as GLP-1(9-36) amide, and no breakdown at the C-terminus was seen.^[4]

Two lessons carry over. First, the product differs from the parent by the loss of the first two residues, His and Ala, so its monoisotopic mass is lower by the sum of their residue masses, 137.059 + 71.037 = 208.096 Da (see the peptide mass calculator). Second, many antibody-based assays recognise the C-terminal part and detect intact and truncated forms together; separating them needed chromatography, which is why HPLC and LC-MS are the usual readouts.^[4]

How the assay is run

Stability assays generally consist of three steps: incubation, precipitation of the matrix proteins, and detection. A systematic comparison found that the methods used for each step differ widely between laboratories, which limits how well results can be compared.^[5] A typical sequence looks like this:

1. Bring the matrix to the incubation temperature, usually 37 °C, and record its source, species, anticoagulant and dilution.
2. Add the peptide from a concentrated stock to a known starting concentration and mix. The dilution calculator helps plan the volumes.
3. Withdraw portions at set times, for example 0, 5, 15, 30, 60 and 120 minutes, and stop the reaction at once.
4. Stop the reaction by precipitating the matrix proteins and centrifuge. In the comparison above, which quantified how much peptide was lost during this step, the widely used precipitation with strong acids was found unsuitable, whereas mixtures of organic solvents preserved more peptide for analysis.^[5]
5. Measure the intact peptide in the supernatant by HPLC or LC-MS, and use mass spectrometry to identify the fragments.

Two controls help. A zero-time sample, quenched immediately after mixing, shows how much peptide survives precipitation, so recovery losses are not mistaken for degradation. A sample in buffer without enzymes separates chemical instability and losses to container walls (see the labware adsorption guide) from proteolysis. The same comparison found strong variation in degradation between plasma and two cell culture supernatants, and compared fluorescent with isotopic labels, so the labelling method should be reported too.^[5]

Turning data into a half-life

If a single enzymatic pathway dominates and the enzyme is not saturated, loss of intact peptide often follows first-order kinetics: ln(C/C₀) = −kt, and the half-life is t½ = ln 2 / k ≈ 0.693 / k.

A worked example with illustrative numbers: if 80% of the intact peptide remains after 10 minutes, k = −ln(0.80) / 10 min = 0.0223 per minute, and t½ = 0.693 / 0.0223 ≈ 31 minutes. Fit k from several time points. A curve that bends on a log plot suggests more than one pathway, enzyme saturation or a recovery problem. Report the matrix, temperature and starting concentration with every value.

Structural features studied for protease resistance

Because exopeptidases attack the ends of a chain, the termini are a common focus. In a series of model peptides, those carrying D-amino acids at both ends were more stable in human serum than predicted from the effect of single substitutions, and N-acetylglucosamine glycosylation stabilised peptides even several residues from the main cleavage site, probably through a change in conformation.^[1] How acetylation, amidation and D-residues are written is explained in the sequence notation guide.

A review of the field covers approaches such as modification of the N- and C-terminus, replacement of L- with D-amino acids, shortening of the chain and conjugation to polymers such as PEG. It cites somatostatin, which persists in plasma for only minutes, against its shortened D-amino acid analogue octreotide at about 1.5 hours.^[6] Albumin conjugates of maleimide-modified GRF(1-29) analogues, the series from which CJC-1295 was selected, showed enhanced in vitro stability against DPP-4.^[7] Semaglutide carries Aib at position 8, the residue at the DPP-4 cleavage site described above, and a fatty-acid side chain that raises its affinity for serum albumin.^[8]

Reading a stability result

A reported half-life describes the peptide and the assay together. Before comparing values, check:

• Matrix. Serum or plasma, which anticoagulant, fresh or stored, which species, and at what percentage.^[3]
• Temperature and time points. Cooling can stop an enzymatic reaction entirely, as it did for DPP-4 in plasma at 4 °C.^[4]
• Recovery. Whether a zero-time sample and a buffer control were run, and how the proteins were precipitated.^[5]
• Identity of products. Whether fragments were identified by mass, as described in the mass spectrometry guide, or only the loss of the parent peak was measured on an HPLC chromatogram.
• Scope. An in vitro half-life ranks peptides under one set of conditions; it does not describe clearance in a living system, which also involves renal filtration.^{[3, 6]}

This guide describes general laboratory methods and published literature only. It is not a statement about the stability or behaviour of any particular material. Terms used here are defined in the glossary, and questions can be sent to support.