Key points
- Mass spectrometry measures mass-to-charge ratio (m/z) of ions. Peptides are ionised gently, by electrospray (ESI) or MALDI, so they reach the analyser intact.
- A single MS measurement confirms the mass. Tandem MS (MS/MS) breaks the peptide along its backbone and reveals its sequence.
- Collision-induced fragmentation mainly cleaves peptide bonds, giving b ions (N-terminal pieces) and y ions (C-terminal pieces). Neighbouring ions in a series differ by exactly one residue mass.
- For any cleavage site, b + y = [M+H]⁺ + 1.007. This simple check identifies complementary pairs in a spectrum.
Getting peptides into the gas phase
A mass spectrometer can only weigh ions in a vacuum. For decades this ruled out large, fragile biomolecules, which decomposed before they could be ionised. Two "soft" ionisation methods solved the problem in the late 1980s, and John Fenn and Koichi Tanaka shared part of the 2002 Nobel Prize in Chemistry for them.
Electrospray ionisation (ESI)
The peptide solution, usually water and acetonitrile with a little formic acid, is sprayed from a fine needle held at a high voltage. The charged droplets shrink as the solvent evaporates until individual peptide ions are released. ESI typically adds several protons to a peptide, producing a series of charge states such as [M+2H]²⁺ and [M+3H]³⁺. Because ESI works directly from a liquid stream, it is easily coupled to HPLC (LC-MS), which is the standard arrangement in proteomics.
MALDI
In matrix-assisted laser desorption/ionisation the peptide is mixed with a large excess of a small, UV-absorbing organic acid, the matrix, and dried onto a metal plate. A laser pulse vaporises the matrix, which carries the peptide into the gas phase and transfers a proton to it. MALDI produces mainly singly charged ions, [M+H]⁺, which makes spectra of mixtures easy to read. Common matrices are α-cyano-4-hydroxycinnamic acid (CHCA) for peptides, sinapinic acid for proteins, and 2,5-dihydroxybenzoic acid (DHB) for glycopeptides and other polar analytes.
| ESI | MALDI | |
|---|---|---|
| Sample form | Solution, continuous flow | Dried spot with matrix |
| Typical charge states | Multiple (2+, 3+ and higher) | Mostly 1+ |
| Coupling to HPLC | Direct, online | Offline, by spotting fractions |
| Tolerance of salts and detergents | Low | Moderate |
| Typical uses | LC-MS/MS proteomics, purity checks, quantitation | Rapid mass checks, peptide mass fingerprinting, imaging |
Mass analysers and resolution
Once formed, ions are separated by m/z in a mass analyser. Time-of-flight (TOF) instruments measure how long ions take to travel a fixed distance; quadrupoles and ion traps filter or store ions using oscillating electric fields; Orbitrap and FT-ICR instruments measure the frequency at which ions oscillate in a trap and convert it into m/z with very high precision.
The key property is resolving power, the ability to separate two close m/z values. Above a resolving power of a few thousand, the individual isotope peaks of a peptide become visible, and the monoisotopic mass can be measured directly. The molecular weight guide explains why this makes the monoisotopic mass the right value to compare with, and how the isotope pattern changes as peptides get larger.
Modern high-resolution instruments measure peptide masses with an accuracy of a few parts per million. For a 1000 Da peptide, 5 ppm corresponds to 0.005 Da, which is enough to rule out most wrong elemental compositions but not to distinguish isomers such as Leu and Ile.
Reading charge states
In ESI a peptide of mass M carrying z protons appears at:
The charge of any peak can be read from its isotope spacing. Isotope peaks are separated by about 1.003 Da in mass, so they appear 1.003 m/z apart for a 1+ ion, 0.50 apart for 2+, and 0.33 apart for 3+. Tryptic peptides, which end in Lys or Arg and have a free N-terminus, most often appear as 2+ ions in ESI.
For Leu-enkephalin (YGGFL, monoisotopic M = 555.2693 Da), [M+H]⁺ appears at m/z 556.2766 and [M+2H]²⁺ at m/z 278.6419.
Tandem MS: fragmenting the backbone
In tandem mass spectrometry one precursor ion is selected, fragmented, and the fragments are measured. The most common fragmentation method is collision-induced dissociation (CID), in which the ion is accelerated into an inert gas such as nitrogen or argon. Its higher-energy variant used in Orbitrap instruments is called HCD. The collisions heat the ion, and the weakest link, which in a protonated peptide is usually a peptide bond, breaks.
When a peptide bond breaks, the ionising proton stays on one of the two pieces. The nomenclature, introduced by Roepstorff and Fohlman in 1984 and refined by Biemann, labels fragments by which bond was broken and which end they contain:
- b ions contain the N-terminus and are numbered by how many residues they contain: b1, b2, b3…
- y ions contain the C-terminus and are numbered the same way from the other end: y1, y2, y3…
- a ions are b ions that have lost carbon monoxide (−27.995 Da). An a2/b2 pair 28 Da apart is a common landmark in CID spectra.
- c and z ions come from cleavage of the N–Cα bond instead of the peptide bond. They dominate in electron-based methods, described below.
For singly charged fragments, the masses follow directly from the residue masses used throughout this site:
Worked example: Leu-enkephalin
Leu-enkephalin is widely used as a calibration and test standard in peptide MS. With the monoisotopic residue masses Tyr 163.06333, Gly 57.02146, Phe 147.06841 and Leu 113.08406, its fragment ions are:
| Bond broken | b ion | m/z | y ion | m/z | b + y |
|---|---|---|---|---|---|
| Y | GGFL | b1 (Y) | 164.0706 | y4 (GGFL) | 393.2132 | 557.2838 |
| YG | GFL | b2 (YG) | 221.0921 | y3 (GFL) | 336.1918 | 557.2839 |
| YGG | FL | b3 (YGG) | 278.1135 | y2 (FL) | 279.1703 | 557.2838 |
| YGGF | L | b4 (YGGF) | 425.1819 | y1 (L) | 132.1019 | 557.2838 |
Every pair adds up to 557.284, which is [M+H]⁺ (556.277) plus one more proton, as expected: both fragments were counted with a proton, but the precursor carried only one. Within each series, consecutive ions differ by a residue mass: b3 − b2 = 57.021 (Gly), b4 − b3 = 147.068 (Phe). Reading these differences along a series is how a sequence is extracted from a spectrum.
Real spectra are less tidy than the table. b1 ions are usually weak or absent. The a4 ion at m/z 397.187 (b4 − CO) is prominent for Leu-enkephalin. Small immonium ions, single residues that have lost CO, appear at the low-mass end and flag which amino acids are present: m/z 136.076 for Tyr, 120.081 for Phe and 86.096 for Leu or Ile.
What makes some fragments stronger
Fragmentation is not random. The mobile proton model explains most of the patterns: cleavage needs a proton on the backbone amide, and basic residues compete for that proton.
- Tryptic peptides give strong y series. The C-terminal Lys or Arg holds a proton, so C-terminal fragments are more likely to keep the charge.
- Cleavage before proline is enhanced. The tertiary amide nitrogen of proline is more basic, and a strong y ion starting at Pro is one of the most reliable features in a CID spectrum.
- Arginine-rich peptides fragment poorly. The guanidinium group holds its proton tightly and leaves little mobile charge for the backbone.
- Neutral losses accompany many fragments: water (−18.011) from Ser, Thr, Asp and Glu; ammonia (−17.027) from Arg, Lys, Asn and Gln; and phosphoric acid (−97.977) from phosphoserine and phosphothreonine.
Electron-based fragmentation
Electron capture dissociation (ECD) and electron transfer dissociation (ETD) fragment peptides by a different mechanism. Adding an electron to a multiply charged peptide produces a radical that cleaves the N–Cα bond, giving c and z ions. The cleavage is fast and does not depend on heating the whole ion, so labile modifications such as phosphorylation and glycosylation stay attached to their residues. ETD also works well for long, highly charged peptides where CID gives poor coverage. Hybrid methods such as EThcD combine both approaches and can even distinguish Leu from Ile through characteristic side-chain losses.
Proline is a blind spot for these methods. Its N–Cα bond is part of a ring, so breaking it does not separate the chain, and no c or z ion forms at proline residues.
From spectra to sequences
In bottom-up proteomics, proteins are digested with trypsin, the peptides are separated by nano-LC and analysed by ESI-MS/MS, and thousands of spectra are recorded per hour. Software such as Mascot, SEQUEST or MaxQuant compares each spectrum with theoretical fragment masses of every tryptic peptide in a protein database. To control false identifications, the search is repeated against a decoy database of reversed or shuffled sequences, and results are filtered to a chosen false discovery rate, commonly 1%.
When no database is available, for example for peptides from venoms or unsequenced organisms, sequences can be read de novo from the mass ladders of b and y ions. This works best with high-resolution data and complementary fragmentation methods.
Limits and pitfalls
- Leu and Ile have identical masses and cannot be distinguished by CID alone. Lys (128.095) and Gln (128.059) differ by only 0.036 Da and need high resolution.
- Some residue combinations mimic single residues. Gly-Gly (114.043) is nearly identical to Asn (114.043), and Gly-Ala or Ala-Gly (128.059) matches Gln. A missing ion in a ladder can make a dipeptide look like one residue.
- Salts, detergents and polymers suppress ESI signals. PEG and Triton contamination produce characteristic ladders 44 Da apart.
- Sodium and potassium adducts appear 21.98 and 37.96 Da above [M+H]⁺ and are easily mistaken for modifications.
- Methionine oxidation during sample handling produces +16 Da peaks that are artefacts, not biology.
Frequently asked questions
What is peptide mass fingerprinting?
An older identification method, usually with MALDI, in which a protein is digested with trypsin and the masses of the resulting peptides are compared with predicted digests of database proteins. It needs no fragmentation, but it works only for relatively pure proteins.
Why do b ions not contain water?
A b ion is formed when the peptide bond breaks and the N-terminal piece keeps the carbonyl group, usually as a cyclic oxazolone structure. It has lost the C-terminal part of the molecule, including the extra water that belongs to the free C-terminus. The y ion keeps that water, which is why its formula includes the extra 18.011 Da.
Can MS tell whether a peptide has a disulfide bond?
Yes. Each disulfide lowers the mass by 2.016 Da, and reduction with a reagent such as DTT restores it. Comparing spectra before and after reduction reveals how many disulfides are present, and fragmentation of partly reduced or digested material shows which cysteines are paired.
How is MS used to quantify peptides?
Most often by adding a known amount of the same peptide labelled with stable isotopes (¹³C, ¹⁵N) and comparing the signal intensities. Because the labelled standard behaves identically but differs in mass, it corrects for losses and ionisation variability.
References
- Fenn JB, Mann M, Meng CK, Wong SF, Whitehouse CM (1989) Electrospray ionization for mass spectrometry of large biomolecules. Science 246:64–71.
- Karas M, Hillenkamp F (1988) Laser desorption ionization of proteins with molecular masses exceeding 10,000 daltons. Analytical Chemistry 60:2299–2301.
- Roepstorff P, Fohlman J (1984) Proposal for a common nomenclature for sequence ions in mass spectra of peptides. Biomedical Mass Spectrometry 11:601.
- Syka JEP, Coon JJ, Schroeder MJ, Shabanowitz J, Hunt DF (2004) Peptide and protein sequence analysis by electron transfer dissociation mass spectrometry. PNAS 101:9528–9533.
- Aebersold R, Mann M (2003) Mass spectrometry-based proteomics. Nature 422:198–207.