Two measurements sit behind almost every peptide certificate. A chromatographic separation reports what share of the detected material is the intended species; a mass measurement reports whether that species carries the expected molecular mass. Published recommendations for synthetic peptides describe exactly this pairing: purity and identity are typically evaluated by analytical reversed-phase chromatography, and mass spectrometry using either electrospray ionisation or matrix-assisted laser desorption/ionisation is treated as essential for identification 1. What follows is an account of how each number is produced, and of the boundary between them.
The separation
Reversed-phase high-performance liquid chromatography separates peptides on a non-polar stationary phase, conventionally alkyl chains bonded to porous silica. Published method descriptions name both common chain lengths: a C8 column of 150 mm × 2.1 mm internal diameter with 5 μm particles, in which the alkyl chains also shield the siloxane bond 2, and a C18 column of the same dimensions used for peptide separations monitored by ultraviolet absorbance and mass spectrometry in series 3.
Elution is by gradient rather than at fixed composition, because peptides in a mixture span a wide range of hydrophobicity. Reported conditions include a linear gradient rising 0.5% acetonitrile per minute at 0.3 mL/min 2, and a longer ramp from 2 to 48% acetonitrile over 86 minutes 3. Retention order and resolution both depend on the slope chosen, so a gradient is part of a method description rather than an incidental setting.
The ion-pairing modifier
A peptide carries positive charges at the side chains of lysine, arginine and histidine and at a free α-amino group, and those charges work against retention on a non-polar surface. An anionic ion-pairing reagent in the mobile phase addresses this directly: negatively charged anions ion-pair with the positively charged residues, which not only neutralises the charged groups, decreasing hydrophilicity, but further increases the affinity of the peptides for the reversed-phase sorbent 2. Trifluoroacetic acid is the conventional choice, with formic acid the usual alternative; both have been characterised across a modifier range of 0.025% to 0.2%, with 0.1% the common working concentration 3.
The choice is not free. Measured side by side, the electrospray signal in the presence of trifluoroacetic acid was suppressed approximately ninefold relative to formic acid, which forces a compromise between the requirements of an effective separation and those of efficient mass-spectrometric detection 3. A certificate that names the modifier is therefore describing a constraint on both measurements at once.
Detection at 214, 220 and 280 nm
Ultraviolet absorbance is the routine detector, and published practice monitors two regions: a backbone wavelength of 214 or 220 nm, and 280 nm 1. Some methods use 215 nm for the backbone channel 3.
The two regions answer different questions because different groups absorb in them. Absorbance in the far ultraviolet arises primarily from the peptide bond itself 4. Absorbance at 280 nm, in contrast, arises strictly from tryptophan and tyrosine residues, and to a small extent from disulfide bonds where present, with the consequence that a peptide containing no tyrosine or tryptophan cannot be quantified at that wavelength at all 4. The difference in magnitude is large: the ratio of absorbance at 205 nm to absorbance at 280 nm averages about thirty across proteins, while varying widely between them 4.
Two consequences follow. A backbone wavelength responds to every amide linkage in the injected sample, so peptide-related impurities — deletion sequences, truncations, oxidised forms — appear in the same denominator as the target. And a figure determined at 280 nm describes only the subset of species carrying an aromatic side chain, which is why the detection wavelength belongs beside any purity percentage.
Electrospray and MALDI
Two ionisation techniques dominate peptide work, and they produce different-looking data from the same molecule.
Electrospray ionisation generates intact ions in vacuo from species in solution, which makes it straightforward to place at the outlet of a chromatographic column. Its distinguishing feature for large molecules is a coherent sequence of peaks whose component ions are multiply charged, the ions of each peak differing by one charge from their neighbours in the sequence 5.
Matrix-assisted laser desorption/ionisation takes a different route, co-crystallising the analyte with a matrix and desorbing it with a laser pulse. For peptides it offers high sensitivity and an excellent tolerance of salt and other common buffer components, with routine detection limits in the subpicomole range; the ions commonly observed are the protonated molecules, which keeps interpretation simple 6.
Charge-state envelopes and deconvolution
The multiply charged series that electrospray produces is the charge-state envelope, and it is not itself a mass. Each observed position relates the neutral mass to the charge as m/z = (M + zH)/z, and charge-state deconvolution transforms the observed m/z values into a neutral, zero-charge mass spectrum 7. The step has a long history in this form: deconvolution of protein charge states has long been used to determine intact mass values from pseudo-molecular ions 8.
Where isotope peaks are resolved, the charge can also be read directly from their spacing, because the gap between adjacent ions in an isotopic envelope is the reciprocal of the charge, and the envelope is characterised by the number of its isotopes and their relative intensities 9.
Deconvolution is an inference, and it has documented failure modes: overlapping charge-state distributions from more than one species in the same spectrum, background noise, and misassignment of harmonic masses that are integer multiples of the true mass 7.
Monoisotopic and average mass
Two different numbers are described as "the mass", and which one an instrument returns depends on whether it resolves the isotope pattern.
The monoisotopic peak is the peak appearing at the theoretical mass, discounting any attached heavy isotopes; it sits alongside the slightly heavier signals contributed by the portion of the sample carrying them 9. Resolving those signals becomes progressively harder as molecular mass rises, since isotope distributions of highly charged ions are obscured by adducts and modifications in the sample 8, and isotopic resolution at chromatographic timescales is already difficult above roughly 25 kDa 7. Where the pattern is not resolved, deconvolution yields average masses rather than the monoisotopic masses obtainable from a high-resolution measurement 8.
A calculated mass on a certificate should therefore state which convention it follows, because comparing an observed average mass with a calculated monoisotopic mass is comparing two different quantities.
What an intact mass does not establish
An intact-mass measurement tests one proposition: that the material has the molecular mass calculated for the intended structure. Reference-standard work illustrates the accuracy available, with an experimental m/z of 1209.6515 against a theoretical 1209.6533 10.
It cannot separate species that share that mass. Isomeric peptides possess identical molecular weights, and stereoisomers in particular present an analytical challenge precisely because of their structural similarity to their all-L counterparts; resolving them required an ion-mobility dimension added to the mass measurement 11. A diastereomer formed by racemisation of a single residue, a transposition of two residues, and the intended peptide are one number to an intact-mass measurement.
Sequence-level confirmation is a separate experiment. In the reference-standard programme it came from tandem mass spectrometry, where fragmentation data yielded complete coverage of the amino acid sequence 10.
A purity figure and an intact mass constrain different things. The chromatogram bounds how much of the detected material is something other than the target; the mass bounds what the target is. Species that are invisible to one are often visible to the other, which is why both appear on a certificate and neither is quoted alone.