Peptide HPLC Sample Preparation: Avoiding Artefacts in QC Runs
When a QC lab tests incoming lots from a large research order, the chromatogram is only as reliable as the solution that went into the instrument. Peptide HPLC sample preparation covers everything done to the material after it leaves its vial and before the autosampler needle draws it: choosing a solvent, weighing, diluting, clarifying and queuing the sample. Each step can add artefacts that look like impurities or hide real ones. This guide walks through the main failure points in the order they happen, with the checks that separate a preparation problem from a problem with the lot itself.
Why peptide HPLC sample preparation matters on multi-lot orders
A single test result can tolerate a little noise. A comparison across lots cannot. If one lot is dissolved on Monday in one solvent and the next is prepared a month later by a different analyst with a different filter, any difference between the two traces is as likely to come from the bench as from the material.
For labs that receive peptides in volume and check each lot on arrival, the answer is a written preparation procedure that fixes the solvent, target concentration, clarification method and maximum hold time. Record any departure in the sample log. That turns preparation from a hidden variable into a controlled one, and it makes lot-to-lot comparisons worth making.
Choosing a solvent for the analytical sample
The first decision is what to dissolve the lyophilised solid in, and the peptide’s charge profile largely decides it:
- Net basic sequences (rich in lysine, arginine or histidine) usually dissolve well in water or dilute acid, and often poorly near neutral or basic pH.
- Net acidic sequences (rich in aspartate or glutamate) tend to resist low pH and go into solution more readily in a mildly basic or buffered diluent.
- Hydrophobic or amphipathic sequences may need a modest fraction of acetonitrile or another organic solvent before they dissolve fully.
Two shortcuts deserve caution. Ultrasonic baths create local hot spots and can generate radicals that oxidise sensitive residues such as methionine and cysteine. Warming the sample speeds up deamidation, cyclisation and hydrolysis. If a solid only dissolved after being pushed hard, treat the resulting impurity profile with some suspicion; the treatment may have made part of it.
Matching sample strength to the starting gradient
This is the error with the most recognisable fingerprint. A reversed-phase gradient normally starts with a low percentage of organic solvent so that analytes stick to the top of the column in a tight band. If the sample itself is dissolved in a solvent richer in organic than that starting condition, the analyte does not bind as it enters. It begins moving down the column while the injection plug is still passing through.
What you see:
- Early-eluting peaks that front or split into two.
- Main peaks that are wider than usual, with poorer resolution from close neighbours.
- Distortion that worsens as the injection volume goes up.
The fix is to dilute the sample into a solvent at or below the starting mobile-phase strength, or to reduce the volume drawn by the autosampler. If a misshapen peak sharpens after that change, the fault was in the sample solution, not in the peptide.
Losses and overloading: two ways to distort a result
Peptides are sticky. Many adsorb to borosilicate glass, some plastics and filter membranes, and the effect is concentration-dependent: trivial in a concentrated stock, potentially large in a dilute working solution.
Container surfaces
Loss to the vial wall lowers the apparent content without any chemical change to the peptide. Worse, different species in the sample may adsorb to different extents, so the ratio between the main peak and a related impurity can shift. Low-adsorption vials, polypropylene inserts, and a small proportion of organic solvent or acid in the diluent all reduce the effect.
Syringe and membrane filters
Filtering protects the column from particulates, but membranes have binding capacity that depends on their chemistry and area. The usual practice is to run the first portion of the filtrate to waste so the binding sites are occupied, then collect the remainder for analysis. Skipping that step on a dilute sample can remove a noticeable fraction of the analyte, and not always evenly across components. When the only aim is to clear particles, centrifuging the sample and drawing from the supernatant avoids membrane contact altogether.
Too much or too little on the column
Too little sample and small impurities sink into baseline noise. Too much and the column’s local capacity is exceeded, which produces fronting or triangular peaks and blurs the separation between the main component and anything eluting just before or after it.
A simple test tells you which side you are on. Halve the amount loaded on the column and repeat the run. If the peak area halves and the shape improves, the original run was overloaded. If the shape stays the same, the broadening comes from somewhere else, such as the solvent mismatch described above, secondary interactions with the stationary phase, or slow conformational exchange within the peptide.
Stability in the autosampler queue
A long sequence of runs can keep prepared samples waiting for many hours. Any degradation route that is fast in solution will progress during that time:
| Feature in the sequence | Change during a long queue | Mass shift |
|---|---|---|
| Free cysteine | Disulfide-linked dimer forms | Roughly double the mass, minus 2 Da |
| N-terminal glutamine | Cyclises to pyroglutamate | -17 Da |
| Asparagine, especially before glycine | Deamidates via a succinimide | +1 Da |
| Methionine | Oxidises to the sulfoxide | +16 Da |
Two controls catch this. Keep the autosampler tray chilled, and bracket the sequence by running the same prepared sample at the start and again at the end. If the bracketing runs disagree, the solution changed while it waited, and the later results in that sequence need to be interpreted with that in mind.
A short troubleshooting routine for unexpected results
When a lot from a volume order produces an odd chromatogram, re-preparation is cheaper than an argument with a supplier. Work through these steps before concluding anything about the material:
- Prepare a fresh sample from a different vial of the same lot.
- Dissolve it in a solvent no stronger than the starting mobile phase.
- Clarify by centrifugation, or discard the first portion of filtrate.
- Load a lower amount and run it promptly on a chilled tray.
- Compare against the original trace and against the lot’s certificate.
If the anomaly disappears, the preparation was the cause. If it persists across vials and preparations, the question returns to the lot, and it is worth contacting the supplier with both traces.
What a certificate will and will not tell you
Certificates of analysis generally list the column, mobile phases and gradient, but rarely say how the sample was dissolved, whether it was filtered, or how long it sat before the run. That is ordinary practice rather than an omission, yet it explains why two competent labs can report slightly different purity for the same lot. Our article on why certificates disagree on purity covers the other contributors.
Bulk Peptides has its products independently tested by HPLC for purity. Certificates for part of the catalogue can be viewed on the certificates of analysis page, and cap and crimp colours link each vial to its report.
Everything described in this article relates to analytical sample handling on laboratory instruments. Our products are sold for in-vitro research only and are not for use in humans or animals.
