Adsorptive Loss: Where the Peptide Goes When It Is Not in the Tube

Two identical tubes will settle the question in an afternoon. Make up the solution, measure it, decant the whole of it into the second tube, let it stand for the same interval, measure again. Should the second number come in lower, no chemistry has gone wrong: material has simply left the liquid and taken up residence on the wall. Almost nobody runs that check. Which is why the phenomenon it exposes gets blamed on degradation, on a short-filled vial, on anything except the container.

One step will do it too, where the sequence happens to carry an absorbing residue. Determine the strength by absorbance, per concentration by A280, and set that figure beside what the balance said should be there. Many sequences carry no such residue, and for those the two-tube comparison remains the practical route.

Fixed in milligrams, variable in percent

Any given container offers a finite population of binding sites. Saturate them and the uptake stops. So the mass that disappears is broadly a constant for one peptide in one vessel, and the fraction that mass represents is whatever the starting load happens to make it.

Load the tube at a milligram per milliliter and a microgram on the wall vanishes into the noise. Load it at a microgram per milliliter and that identical microgram is a serious bite out of the sample. The arithmetic runs backwards from intuition, and it puts the failure precisely where suspicion is weakest — not in the concentrated stock, which is robust, but in the dilute working standard, which is not.

Chains of dilution stack the effect. Every vessel along the chain claims its portion, each portion representing a bigger slice than the last, so a nominally correct final solution can be meaningfully short while not one intermediate gave any sign of trouble.

Several grips at once, and the surface picks which

Molecules accumulating at an interface is adsorption in general; peptides excel at it because a single chain offers a surface more than one way to hold on. Ionic attraction operates through charged side chains. Hydrophobic residues tuck against hydrophobic material. Hydrogen bonding runs underneath all of it.

Which grip prevails is a property of the material, not of the peptide alone.

Move a well-behaved peptide from plastic into glass and it may misbehave; move a troublesome one the other way and it may settle down. No container is correct for everything, and that is the reason.

Surfaces that never make it into the count

Pipette tips are the quiet one. Contact is brief, but the area is enormous against the volume held, and every tip is a clean, wholly unsaturated surface. Because tips get swapped continually, the loss repeats instead of exhausting itself, riding on top of the volumetric error treated in pipetting accuracy and volumetric error. Autosampler vials and inserts contribute differently: a dilute sample queued for hours has ample time to reach equilibrium with its walls, which is among the several things unfolding in the gap between the vial and the autosampler. Filter membranes are built for maximum area, and area binds. Then there is every intermediate transfer vessel in the preparation, each one a further surface nobody wrote down.

When the composition moves, not just the strength

Losing everything evenly alters a concentration. Losing some things faster than others alters a composition, and that belongs to a separate category of trouble.

Suppose the target binds harder than one of the impurities travelling with it. The impurity is then concentrated in what remains dissolved, and purity assessed on that liquid comes out low against the truth. Flip the relative affinities and the same measurement flatters the material instead. Aggregate species thrown up by the agitation covered in vortexing, sonication and shear follow the identical logic. Nothing about this registers as an anomaly: peak ratios are wrong while the chromatogram itself looks entirely ordinary, so the fault stays out of sight unless a recovery check goes looking.

Countermeasures, least structural to most

• Consumables sold as low-bind. The surface treatment cuts ionic and hydrophobic interaction together. It does not abolish either, and at genuinely low concentrations it will not stand in for the measures further down this list.
• Acetonitrile in the diluent. Five to ten percent is enough to take a large bite out of hydrophobic adsorption, assuming the downstream method tolerates the solvent.
• Working away from the isoelectric point. Net charge shifts, and with it the ionic share of the binding — alongside the solubility consequences set out in isoelectric point and solubility.
• Carrier protein or non-ionic surfactant. An unrelated protein, or a trace of surfactant, occupies the sites so the analyte need not. It works, and it puts something else in the sample, which suits bioassay work far better than analytical work.
• Sacrificial rinse. Fill the vessel with a portion of the same solution, throw it away, then add the real sample to an already-saturated surface. The reasoning matches discarding the leading portion of a filtrate.
• Fewer containers. Build the solution in whatever vessel the measurement will be taken from, and let the intermediates disappear. Nothing else on this list removes the problem rather than shrinking it.

Reading this back onto a certificate

Every number a certificate reports was taken from a solution somebody had to prepare, and these surfaces were present for that too. Laboratories in good order handle it through saturation and a short transfer chain, and whatever residual error survives is absorbed into the uncertainty discussed in measurement uncertainty. So this is not a certificate problem.

It does account for something reported often enough to be worth naming: credible paperwork arrives with the material, and the material nonetheless seems thin in dilute use. Underfilling and degradation are the expensive hypotheses. The cheap one is that a share of the peptide never made it out of the tube where the dilution was done.