Peptide mass spectrometry: how ESI, MALDI, MS/MS and LC-MS confirm a peptide's identity

On most peptide certificates, mass spectrometry gets one line: “mass conforms”. Behind that line is a measurement with several choices that decide whether it actually proves anything: which ionisation method, which charge states, which theoretical mass, what resolution and what tolerance. This guide explains each of them, then goes further into MS/MS sequencing, peptide mapping and LC-MS impurity work.

It goes deeper than our overview of peptide purity testing, which explains where MS sits next to HPLC, and our guide on how to read a peptide COA.

What does a mass spectrometer measure?

A mass spectrometer does not weigh molecules directly. It turns them into gas-phase ions and measures each ion’s mass-to-charge ratio (m/z). For a peptide carrying z extra protons, the observed m/z is the peptide’s mass plus z protons, divided by z. Everything else on an MS report, including the “molecular mass” figure, is derived from those m/z values.

EMA describes MS as “a powerful analytical tool for the structure elucidation of peptides” whose variants “can be used to determine the molecular mass of a peptide and to confirm its amino acid sequence” [1]. Those are two different experiments: intact-mass measurement, and tandem MS (MS/MS) of fragments.

ESI vs MALDI: how peptides are ionised

Peptides are large, polar and fragile, and before the late 1980s they were hard to get into the gas phase intact. Two “soft” ionisation methods solved that.

Electrospray ionisation (ESI) produces “intact ions in vacuo from large and complex species in solution” [2]. A charged liquid is sprayed from a capillary at high potential; the droplets shrink by solvent evaporation and Coulomb explosions until gas-phase ions are released, by ion evaporation or charge-residue mechanisms [3]. Because it works from a flowing liquid, ESI couples directly to HPLC. Kebarle and Verkerk’s review also covers how the ESI mechanism affects protein spectra and the role of additives such as ammonium acetate on the observed spectra [3], one reason the same peptide can show a different charge-state distribution on two instruments or in two mobile phases.

Matrix-assisted laser desorption/ionisation (MALDI) is the second method. In 1988, Karas and Hillenkamp reported laser desorption ionisation of proteins with molecular masses exceeding 10,000 daltons [4]. The sample is mixed with a light-absorbing matrix, dried on a plate and desorbed by a laser pulse, usually into a time-of-flight (TOF) analyser, the MALDI-TOF format Medzihradszky and Chalkley describe [5].

	ESI	MALDI
Sample state	Liquid, often straight from LC	Dried with a matrix on a target plate
Typical ions	Series of multiply charged ions	Ions desorbed by a laser pulse, usually analysed by TOF
Coupling	Online LC-MS	Usually offline
Strength for peptides	LC-MS impurity work, MS/MS sequencing, quantitative LC-MS	Fast intact-mass checks
Ref	[2], [3]	[5]

Medzihradszky and Chalkley summarise current practice: MALDI-TOF “is the easier-to-use technique”, but “ESI coupled with on-line fractionation methods is the most widely used peptide analytical technique” [5].

Charge states and deconvolution

The defining feature of ESI spectra, in Fenn’s words, is “coherent sequences of peaks whose component ions are multiply charged, the ions of each peak differing by one charge from those of adjacent neighbors in the sequence” [2]. One peptide therefore appears as several peaks, and software combines them (deconvolution) into a single neutral mass.

A worked example shows why the raw spectrum looks the way it does. Using tirzepatide’s monoisotopic mass of 4810.5249 Da [6] and a proton mass of 1.0073 Da, the monoisotopic m/z values are:

Charge state	Ion	Calculated monoisotopic m/z	Spacing between isotope peaks
3+	[M+3H]³⁺	1604.52	0.33
4+	[M+4H]⁴⁺	1203.64	0.25
5+	[M+5H]⁵⁺	963.11	0.20
6+	[M+6H]⁶⁺	802.76	0.17

These are calculated values for illustration, not measured data. Two practical points follow. First, the charge state can be read from the isotope spacing: peaks 0.25 m/z apart mean z = 4. Second, the higher the charge, the closer the isotope peaks, so resolving them takes more resolving power.

Monoisotopic vs average mass: which number should match?

This is where many certificates go wrong. A peptide does not have one mass. It has a distribution of masses, because each element exists as several stable isotopes. Carbon is the main contributor: CIAAW gives the amount fraction of ¹³C in natural carbon as 0.0096–0.0116 [7].

• Monoisotopic mass is the mass of the molecule made entirely of the lightest isotope of each element (¹²C, ¹H, ¹⁴N, ¹⁶O, ³²S). It corresponds to the first peak of a resolved isotope pattern.
• Average mass (the molecular weight in g/mol) is weighted by natural isotopic abundance. It corresponds to the centre of the isotope envelope and is what you use to weigh out moles.

For tirzepatide, C225H348N48O68, PubChem lists a monoisotopic mass of 4810.5249 Da [6], while its average molecular weight is about 4813 g/mol [6], a difference of about 3 Da. PubChem also lists a third figure, 4812.5316, under “exact mass” [6]; it is not the monoisotopic mass, and comparing an observed monoisotopic peak with it would create a false 2 Da error.

Why does the monoisotopic peak sometimes seem to be missing? Considering carbon alone, the chance that all 225 carbons in tirzepatide are ¹²C is about 0.989²²⁵, roughly 0.08, using a ¹³C fraction within the CIAAW interval [7]. So the monoisotopic peak is a small shoulder at the light end of the envelope, and the tallest peak is several daltons heavier. Senko, Beu and McLafferty described exactly this problem for large biomolecules: the low probability of the all-monoisotopic species creates “the potential to misassign the number of heavy isotopes in a specific peak and make a mass error of ±1 Da”, and they fitted the measured isotope distribution to a model to assign the monoisotopic mass “even in cases where the monoisotopic peak is absent from the spectrum” [8].

For a small peptide the problem shrinks. Linaclotide, C59H79N15O21S6, has a monoisotopic mass of 1525.3899 Da [9] against an average molecular weight of 1526.8 g/mol [9], and with 59 carbons the all-¹²C species is roughly half of the carbon distribution by the same arithmetic.

The rule for reading a report: a resolved isotope pattern should be compared with the monoisotopic mass; an unresolved envelope or a low-resolution deconvolution should be compared with the average mass. The certificate should say which it used.

Worked example: what MS can and cannot say about a disulfide peptide

Linaclotide is a 14-residue peptide cyclised by three disulfide bridges. Each disulfide bond means two cysteine thiols have each lost a hydrogen atom, so the fully reduced linear form is six hydrogen atoms heavier than the folded molecule. From the monoisotopic mass of 1525.3899 Da [9], the reduced form would sit at about 1531.44 Da, a calculated shift of +6.05 Da. Intact MS therefore separates reduced, partly reduced and fully oxidised forms easily.

What intact MS cannot do is tell you which cysteines are paired. Every arrangement of three disulfides in the same sequence has the same mass. Yet mispairing is a real risk: a free cysteine “may cause unwanted intramolecular disulfide scrambling”, and most disulfide degradation is observed under neutral and basic conditions [10]. Establishing the pairing needs fragmentation that crosses the disulfide bonds, partial reduction and alkylation, or mapping of non-reduced fragments. Electron capture is relevant here: Zubarev and colleagues proposed that the hydrogen atom released on capture is taken up at high-affinity sites “such as -S-S- or backbone amide” [11]. EMA asks for characterisation of disulfide bridges and cyclic structures where relevant [1].

High-resolution mass spectrometry: Orbitrap, FT-ICR and TOF

Resolution (how well neighbouring peaks are separated) and mass accuracy (how close the measured value is to the true one) are what make an MS identity test specific.

• Orbitrap. Makarov’s electrostatic trap, in which ions orbit a central electrode with an axial frequency proportional to (m/z)^−1/2 and are detected by image current and Fourier transform, demonstrated mass resolution “up to 150,000” in its first description [12].
• FT-ICR. Fourier transform ion cyclotron resonance traps ions in a magnetic field; Marshall and colleagues’ primer covers how its mass resolving power and mass accuracy arise [13].
• TOF. Time-of-flight analysers, used with MALDI and in Q-TOF instruments, are another common option [5].

Medzihradszky and Chalkley note that “numerous mass spectrometers” now “afford mass accuracy within a few ppm”, and that Gln and Lys, which differ by 36 mmu, can be distinguished if fragment masses are measured accurately enough [5]. ICH Q2(R2) recognises the same principle: specificity can sometimes be justified by technical parameters such as “resolution of isotopes in mass spectrometry” without additional experiments [14].

MS/MS fragmentation: b and y ions, CID, HCD and ETD

Intact mass confirms composition, not order. Two peptides with the same residues in a different order have the same mass. Sequence comes from tandem MS: the peptide ion is isolated, fragmented and the fragments measured.

The nomenclature. The first common nomenclature for peptide sequence ions was proposed by Roepstorff and Fohlman in 1984 [15], and the terminology now used was established by Biemann [5]. Backbone cleavage can occur at three bonds in each residue; “a, b, and c fragments are formed when the charge is retained at the N-terminus, while x, y, and z ions are produced with C-terminal charge retention”, numbered from their own terminus [5]. Because a b ion and its partner y ion come from the same broken bond, “b(i) + y(n−i) = MH⁺ + 1” [5]. The mass difference between neighbouring ions in one series is one amino acid residue, which is how the sequence is read.

Activation method	How energy is added	Main fragments	Notes	Ref
CID (collision-induced dissociation)	Collisions with neutral gas	b and y	Most widely used; low-energy CID in ion traps and collision cells	[16], [5]
HCD (higher-energy collisional dissociation)	Beam-type collisions in a dedicated cell, then Orbitrap detection	b and y, plus immonium ions	Full mass range, high-accuracy fragments; immonium ions help pinpoint modifications	[17], [5]
ETD (electron transfer dissociation)	Electron transferred from an anion to a multiply charged peptide	c and z•	Backbone cleavage analogous to ECD; first demonstrated on phosphopeptides separated by nanoflow HPLC	[18], [5]
ECD (electron capture dissociation)	Low-energy electrons captured by multiply charged ions	c and z•	Many more backbone cleavages than CID; modifications less easily lost	[11]

Some details matter when you interpret spectra:

• In collision-cell CID and HCD, “b-type fragments are structurally less stable than y ions”, so large b ions break down further into smaller b, internal and immonium ions [5].
• ETD and ECD give “almost exclusively peptide backbone fragmentation via cleavages between the amino groups and the alpha carbons”, and their spectra “are complementary to CID data” [5]. In ECD, capture is proposed to release a hydrogen atom that is captured at a high-affinity site “such as -S-S- or backbone amide” [11], which is one reason electron-based methods are attractive for disulfide-containing peptides.
• In Olsen and colleagues’ HCD work, immonium ions pinpointed modifications such as phosphotyrosine “with very high confidence” [17].
• With ordinary low-energy MS/MS, “one cannot differentiate between isomeric Leu and Ile”; high-energy CID can, through side-chain fragments [5].

Biemann’s 1988 review already listed “confirmation of the structure of synthetic peptides” among the core uses of MS/MS, alongside locating modifications and detecting amino acid replacements [19].

Peptide mapping: sequence and modification sites for longer peptides

For longer peptides, MS/MS of the intact molecule may not cover every bond. Peptide mapping cleaves the molecule, usually with an enzyme, and analyses the fragments by LC-MS. EMA says peptide mapping “may be applicable for longer peptides based on cleavage site(s) in the primary structure in cases where the MS/MS sequencing data may be difficult”, and points to Ph. Eur. general chapter 2.2.55 [1].

Mapping is also how modifications are located. Ji and colleagues used liquid chromatography, peptide mapping and MS to identify and quantify which methionine and tryptophan residues of PTH(1-34) oxidised under different stresses [20]. Chelius and colleagues used tryptic peptide mapping and MS/MS to identify N-terminal pyroglutamate in antibodies [21]. For disulfide peptides, comparing non-reduced and reduced maps shows which cysteines are paired, a characterisation EMA asks for where relevant [1].

The harmonised chromatography chapter allows visual comparison of profiles as a system suitability test for complex chromatograms of biotechnological and biological products, and the Japanese Pharmacopoeia adds, as a local provision, that for peptide mapping the separation pattern “may be set for acceptance criteria as a profile” [22]. The whole trace, not a single number, is compared with the reference.

LC-MS impurity identification

When a new peak appears in an HPLC chromatogram, LC-MS is how it gets a name. The mass difference from the parent points to the chemistry: a missing residue for a deletion sequence, +16 Da for oxidation, about +1 Da for deamidation, the mass of a protecting group for incomplete deprotection [23]. Our guide to peptide degradation pathways lists the shifts for each degradation route.

How far this can go is shown by metrology studies done for international comparisons of peptide purity:

• In a synthetic human C-peptide study material, LC-hrMS identified, confirmed and quantified “more than 65 impurities”, totalling 83.3 mg/g [24].
• In a synthetic oxytocin material, eighteen structurally related impurities were quantified, totalling 31.1 mg/g [25].
• In a “pure” angiotensin I material, the five major impurities, at 0.66 to 4.86 mg/g, included three angiotensin I isomers with leucine/isoleucine swaps [26].

The last example is the warning. Isomers have the same mass, so they had to be separated by LC and quantified against original impurity standards [26]. Mass spectrometry identifies what the chromatography has already separated.

What a mass spec report on a COA should show

EMA’s expectation for synthetic peptide characterisation is specific: “Typical representative spectra and interpretation of the fragmentation data, including assignments and tables with theoretical and observed mass values, should be provided” [1]. A batch certificate is shorter than a regulatory dossier, but the same elements make an MS result checkable:

Item on the report	Why it matters
Ionisation and analyser (ESI-Q-TOF, ESI-Orbitrap, MALDI-TOF)	Sets the resolution and accuracy you can expect
Observed m/z values and assigned charge states	Lets you recompute the mass yourself
Deconvoluted or monoisotopic observed mass	The figure being compared
Theoretical mass, stated as monoisotopic or average	Avoids the 1–3 Da mismatch described above
Mass error (ppm or Da) and acceptance tolerance	Shows whether “conforms” is meaningful
MS/MS sequence coverage, if claimed	Confirms order, not just composition
Batch number on the spectrum	Links the data to your vial

Identity should not rest on one technique. EMA recommends “at least two orthogonal methods” for identification at release, naming mass, relative retention time, LC-MS, peptide mapping, bioactivity, amino acid analysis and NMR as appropriate options [1].

What mass spectrometry cannot tell you

• Stereochemistry and isomerism. D-residues, isoaspartate and Leu/Ile swaps leave mass unchanged [5], [26]. These need chromatographic or chiral methods.
• Purity by itself. Peak intensity depends on ionisation efficiency, so MS is used to identify peaks while HPLC-UV quantifies them; see peptide purity testing.
• How much peptide is in the vial. Water, counter-ions and salts are not part of the intact-mass measurement. Content needs a separate measurement; see net peptide content and amino acid analysis.

How Winhycare Pharma supplies research peptides

Winhycare Pharma supplies research peptides from Hyderabad, India, as research-use-only material, at purity of ≥99% as stated in our product brochure. Tirzepatide, semaglutide, retatrutide, cagrilintide and BPC-157 are also supplied as API to licensed manufacturers; see tirzepatide API and peptide APIs. For new sequences we offer custom peptide synthesis. Ask for the batch Certificate of Analysis, including the mass spectrum, with your quote: request research peptides.

For laboratory research use only. Not for human or veterinary use.