Peptide Laboratory Practice
Peptide LC-MS Adducts and Deconvolution Errors
Diagnose sodium, potassium and TFA adducts in peptide LC-MS, avoid false molecular-weight failures, and audit deconvolution settings before rejecting a batch.
The most common peptide LC-MS dispute we see begins with two screenshots. The supplier's report shows the expected neutral mass; the receiving laboratory sees a cluster of peaks 22 or 38 Da away and calls the lot incorrect. Sometimes the material is wrong. More often, the two laboratories used different solvents, source settings, charge-state ranges, or deconvolution rules. Adduct chemistry was converted into an identity failure before anyone inspected the raw spectrum.
Peptides ionized by positive-mode electrospray usually appear as several protonated charge states rather than one molecular ion. Sodium and potassium can replace a proton, trifluoroacetate can remain associated, and oxidation or water loss can add further satellites. A deconvolution algorithm must group those observed ions into a neutral mass. If it groups the wrong envelope, accepts noise, or assumes the wrong adduct, the clean number on the report may be confidently wrong.
Read the charge envelope before the deconvoluted result
For a peptide of neutral mass M and charge z, the protonated ion appears near (M + zH)/z. Adjacent charge states should reconstruct to the same neutral mass. That relationship is the first identity check. Do not begin with the deconvolution pane; open the centroid or profile spectrum across the chromatographic peak and label the charge states manually.
A sodium substitution produces an approximately 21.982 Da shift at neutral-mass level relative to protonation, while potassium produces about 37.956 Da. On the observed m/z axis, the separation is divided by charge. For a 4+ ion, a single sodium adduct sits only about 5.5 m/z units from its protonated partner. Analysts who compare the observed spacing directly with neutral-mass differences often assign an impossible impurity.
TFA creates a different nuisance. It is widely used in peptide purification and may persist as counterion after lyophilization. In negative mode or under certain source conditions, TFA-related species are conspicuous; in positive mode it can suppress response and complicate associated-ion patterns. Acetate, formate, chloride, phosphate, and residual metal salts contribute their own signatures. The sample preparation record is therefore part of the mass-spectrum interpretation, not administrative paperwork.
Why deconvolution returns the wrong molecular weight
Automated deconvolution is sensitive to the selected chromatographic window. Integrating across a broad region can combine the main peptide with a nearby impurity. Background ions then become candidate charge states. A narrow window centered on the UV or extracted-ion apex, followed by scans from the leading and trailing edges, usually reveals whether one envelope belongs to one chromatographic component.
The configured mass range and charge range matter just as much. Excluding the dominant charge state can force the software to build a neutral mass from weak ions. Excessively broad ranges invite unrelated clusters. Smoothing, baseline subtraction, minimum intensity, isotope model, and allowed adducts should be recorded in the method. “Processed by vendor software” is not enough to reproduce a COA.
We also see average mass compared with monoisotopic mass. For a multi-kilodalton peptide, the numerical difference is not trivial. The theoretical value must state whether it represents the neutral monoisotopic molecule, average molecular weight, a salt, or a solvated form. Modifications—amidation, N-terminal acetylation, disulfide pairing, lipidation, PEG-like linkers, or metal complexes—must be included explicitly.
Before rejecting a lot, run this compact audit:
- inspect raw spectra at the peak apex and confirm at least two consistent charge states;
- calculate protonated, sodium-adducted, and potassium-adducted m/z values at the observed charges;
- compare theoretical and reported values on the same monoisotopic or average-mass basis;
- reprocess with a justified mass and charge range, preserving screenshots of every setting;
- prepare a fresh dilution with LC-MS-grade water and organic solvent in a low-binding vial;
- inject a blank after the sample to separate carryover from a genuine peptide-related ion.
Reduce adducts without erasing evidence
Desalting can improve a spectrum, but it should not be used silently to make a failed result pass. Analyze the original preparation first, then repeat after a documented cleanup. A C18 tip or small SPE cartridge may remove nonvolatile salts; recovery must be checked because hydrophilic peptides can pass through and strongly hydrophobic peptides may not elute completely. Changing from TFA to formic acid can increase MS response, yet it may also change chromatographic selectivity.
Source conditions need restraint. Raising cone voltage or in-source collision energy can strip weak adducts, but it can also produce fragments that deconvolution mistakes for intact material. Tune on a suitable reference and retain a low-energy acquisition for identity. When metal adduction persists, inspect glassware, mobile-phase preparation, and shared LC plumbing before blaming synthesis.
Storage artifacts are often misread as adduct problems. Oxidation adds about 15.995 Da, deamidation roughly 0.984 Da, and dehydration subtracts about 18.011 Da at neutral level. Those are chemical mass changes and may show different retention, whereas a labile sodium association can rise or fall with solvent and source conditions. Time-course reinjection helps distinguish the two.
What to demand from a supplier report
A useful LC-MS identity package contains the raw charge-state spectrum, the chromatographic extraction window, theoretical mass calculation, processing parameters, and the deconvoluted output. It identifies the ionization mode and mobile-phase additives. A single cropped peak labeled “M+H” for a 4 kDa peptide is usually a deconvoluted value mislabeled as an observed ion.
For modified peptides, ask for the exact structure used in the calculation. A product name alone does not resolve whether a linker, fatty acid, counterion, or terminal group is present. If multiple LC-MS methods disagree, compare raw ions under harmonized preparation before averaging numbers.
LC-MS is powerful because several independent observations—retention, isotope pattern, charge envelope, accurate mass, and fragments—can converge. Deconvolution should summarize that evidence, not replace it. Peptides Archive can assist with raw-data review and method-transfer questions for RUO peptide lots. This information is for analytical research only and does not support human administration or medical use.
Primary records and verification routes
Use the primary paper, current regulator record, or lot-linked analytical file for the claim it supports. A search result is a route to evidence, not evidence itself.