Mass spectrometry as identity evidence

A mass spectrometry entry on a certificate of analysis is usually two numbers and a verdict. A theoretical mass, an observed mass, and a note that they agree. It is the shortest line on the page, and it is often the one an analyst reads fastest, because it looks like arithmetic. Two numbers match or they do not.

That reading is wrong in both directions. A mass that matches is weaker evidence of identity than it appears to be, because many different molecules share a mass. A mass that does not match is frequently not a problem with the material at all, but a mismatch between two conventions for stating what a mass is. Most of the value in this line comes from knowing which of those situations you are in, and the certificate does not always tell you.

What the instrument measured

A mass spectrometer does not weigh the substance in the container. It measures the mass-to-charge ratio of ions, and it measures it on whatever species survived sample handling, chromatography if any, and ionisation. Every complication in this article follows from that sentence. The pharmacopoeial framing is the same: USP <736> defines the technique in terms of mass-to-charge ratio of ionic species, and Ph. Eur. 2.2.43 treats the subject on the same basis.

The theoretical mass is not measured at all. It is computed from a proposed molecular formula, which is itself derived from a claimed sequence and a set of assumptions about termini, modifications and protonation. So the comparison on the certificate is between a measurement and a hypothesis. When they agree, what has been established is that the sample contains an ionisable species whose mass is consistent with the hypothesis. That is a real result. It is not the same as confirming the hypothesis.

Monoisotopic and average mass are different numbers

This is the most common source of manufactured discrepancies, and it is entirely avoidable.

Average mass is weighted across the natural isotopic abundance of every element in the formula. Monoisotopic mass is the sum of the most abundant isotope of each element, and corresponds to the first peak in the isotope cluster. For a small molecule the two figures nearly coincide. For a peptide, they diverge, and the divergence grows with the number of carbon atoms. On a material of a few thousand daltons the gap is comfortably larger than the tolerance most certificates apply.

Compare a monoisotopic observation against an average theoretical value and you will see a difference that has nothing to do with the material. Compare them the other way and a genuine small difference can be swallowed. Neither error announces itself. IUPAC's mass spectrometry terminology recommendations define both quantities explicitly, which is a useful reference precisely because certificates so often use the words loosely.

There is a second question hiding behind the first: whether the instrument had the resolving power to justify a monoisotopic assignment. A low-resolution instrument on a larger peptide does not resolve the isotope envelope. It reports a centroid of the unresolved cluster, which approximates the average mass. A monoisotopic figure quoted from such a run is a label, not a measurement.

A matching mass does not identify a sequence

Intact mass is a composition check. Composition does not fix arrangement, and for peptides the space of things that share a composition is large.

• Sequence isomers. Any permutation of the same residues has the same formula and therefore the same exact mass, at any resolving power. A transposition of two residues is invisible to an intact mass measurement.
• Leucine and isoleucine. Identical composition. No mass difference exists to measure.
• Lysine and glutamine. These differ by roughly 0.036 u, which is resolvable on a high-resolution instrument and invisible on a low-resolution one.
• Stereochemistry. D- and L-forms have identical mass. Racemisation at one centre is entirely mass-silent.
• Deamidation. Conversion of an amide residue to the corresponding acid shifts the mass by about one unit, which sits close to isotope spacing and is easy to miss in a poorly resolved spectrum.
• Disulfide arrangement. Forming a disulfide bond changes mass by about two units, so the number of bonds is visible. Which cysteines are paired with which is not.

What does discriminate: tandem MS with adequate fragment coverage across the sequence, peptide mapping after enzymatic digestion, amino acid analysis, chiral methods for stereochemistry, and co-elution against a characterised reference under an orthogonal separation. USP <1055> covers peptide mapping as a technique; USP <1503> discusses the attributes that describe a synthetic peptide substance and makes clear that identity is established by a set of methods rather than one. ICH Q2(R2) frames the same point as specificity: an identification procedure has to discriminate the analyte from things it might be confused with.

Adducts, salt forms and charge states

The raw number the detector produces is a mass-to-charge ratio, so the charge state has to be assigned before any mass can be quoted. Electrospray typically produces a series of multiply charged ions. Software deconvolutes that series into a neutral mass, which means the number printed on the certificate is a computed quantity derived from a model of the charge envelope. Deconvolution is usually reliable and occasionally not, particularly on mixtures or on poor signal.

Adducts move the number too. A sodium ion in place of a proton shifts the apparent mass by roughly twenty-two units. Potassium and ammonium adducts, solvent adducts and residual mobile-phase modifiers each produce their own satellite peaks. An analyst who reads the tallest peak without checking the assignment can arrive at a mass that is off by a clean, plausible-looking increment.

Salt form deserves separate attention, because it is a frequent over-read. A peptide supplied as a salt generally ionises as the peptide species. The counterion does not travel with it into the reported mass. That means a mass spectrometry result cannot tell you the salt form, cannot tell you the counterion content, and cannot substitute for a method that measures either. A certificate that appears to confirm a salt form on the strength of a mass number is asserting more than the measurement supports.

The ionisation method is part of the result

Two laboratories can report different numbers for the same material and both be correct. The ionisation method usually explains it. MALDI commonly yields singly charged ions, and on a linear time-of-flight instrument the figure reported tends to be an average mass; a reflectron on a smaller peptide can resolve isotopes and support a monoisotopic figure. Electrospray yields the multiply charged series described above. Matrix choice, mobile-phase composition and source conditions all change which adducts appear, and harsh source conditions can produce in-source fragmentation that populates the spectrum with species the vial never contained.

So when your own re-measurement disagrees with an incoming certificate, method difference is the first hypothesis, not material difference. USP <1736>, the informational chapter that accompanies <736>, exists partly because these application details determine what a result means.

Reading the entry

1. Establish the convention on both sides. Monoisotopic against monoisotopic, average against average. If the certificate does not state which, the comparison is not yet a comparison.
2. Check the charge state and adduct assignment behind the reported mass, not just the final figure.
3. Ask whether the instrument's resolving power supports the number of decimal places quoted.
4. Confirm whether the theoretical value refers to the free peptide or to some other form, and on what assumptions about termini and modifications.
5. Read a match as "consistent with", and record it that way in your own incoming inspection.
6. Where identity is critical, require an orthogonal method and a comparison against a characterised reference. One number is not a characterisation.

None of the documents referenced here are cited as a statement that any particular material conforms to them. They are cited because they define the vocabulary the certificate is using, and because that vocabulary is where the ambiguity lives.

What a mass spectrometry result establishes, stated narrowly: under the stated ionisation and acquisition conditions, the sample produced ions whose mass is consistent with the proposed composition. Everything past that is inference, and inference belongs to the analyst reading the page, not to the instrument.