ESI And MALDI: Two Ways To Weigh A Peptide

Both techniques answer the same question: what is the mass-to-charge ratio of the ions produced from this sample. What differs is the transfer step. In ESI a solution is pushed through a fine capillary held at a potential of roughly 2 to 5 kV against a counter-electrode, the emerging liquid breaks into a spray of charged droplets, solvent evaporates, and bare peptide ions are left in the gas phase. In MALDI the sample is mixed with a large excess of a small ultraviolet-absorbing organic acid, dried on a metal target, and hit with a pulsed laser, historically a 337 nm nitrogen laser and now more often a 355 nm frequency-tripled Nd:YAG at repetition rates of several hundred to a few thousand pulses per second.

The charge distribution is the practical difference. A 4 kDa peptide in positive-mode ESI commonly appears as a family of peaks at charge states from 2+ to 5+, each at a different mass-to-charge ratio, and the mass is reconstructed from the spacing between them. The same peptide in MALDI usually appears as a single protonated species. One spectrum is a puzzle with redundancy built in; the other is a single line that has to be right first time.

Neither technique measures purity. Ionisation efficiency is a property of the individual molecule, not of how much of it is present, so two species in the same vial can differ in signal by an order of magnitude at equal mass. Purity figures come from chromatography.

The two ionisation routes side by side

Mass accuracy expressed in parts per million is only meaningful with the calibration method attached, which is why the table separates the two rows. A 10 ppm figure on a 4,000 Da peptide is 0.04 Da; the same instrument run against an external calibration file recorded hours earlier can drift by several times that.

Matrices, and why the choice is not cosmetic

MALDI depends on a matrix that absorbs at the laser wavelength and co-crystallises with the analyte at a molar excess of roughly 1,000:1 to 10,000:1. Alpha-cyano-4-hydroxycinnamic acid is the usual choice below about 10 kDa and produces relatively little fragmentation. 2,5-dihydroxybenzoic acid gives coarser crystals and tolerates more salt. Sinapinic acid is used higher up the mass range. The matrix is present in vast excess, so its own cluster ions dominate everything below roughly 500 Da and a small peptide can be lost inside that region entirely.

Crystallisation is also uneven. Dried-droplet preparations concentrate analyte at the rim of the spot, so signal intensity varies across a single sample position and spectra are normally summed from several hundred to a few thousand laser shots at different points to average that out.

The failure mode: trifluoroacetate suppression in ESI

Synthetic peptides purified by reverse-phase chromatography usually leave that process as trifluoroacetate salts, and residual trifluoroacetic acid is the single most reliable way to lose an ESI signal. Trifluoroacetate is a strong ion-pairing agent: it binds the peptide's basic sites, resists evaporation from the droplet, and suppresses the formation of the higher charge states that make the spectrum readable. The visible result is not a wrong mass. It is a spectrum with a collapsed charge envelope, a poor signal-to-noise ratio, and a series of adduct peaks at intervals of 114.03 Da, which is the neutral mass of trifluoroacetic acid.

The counter-measure is chemical, not instrumental: adding a weak acid such as 0.1 percent formic acid, or a small quantity of propionic acid, displaces the trifluoroacetate ion pair. A failing spectrum that recovers on changing the mobile phase additive was never an instrument problem. The same residue is close to invisible to MALDI, which is one reason the two techniques are often run on the same material.

Which record answers which question

• Confirming that the observed mass matches the mass calculated from the sequence: either technique, provided the calculated value and the measured value are the same kind of mass. A monoisotopic measurement compared against an average calculated mass will appear to disagree by several Daltons on a peptide of a few kilodaltons.
• Separating and identifying a co-eluting impurity: ESI, because it runs in line with the chromatography and reports a mass for each point in the elution.
• Screening many spots quickly with tolerance for buffer salts: MALDI, on a dried target.
• Locating where on the chain a modification sits: neither, on its own. That requires fragmentation of the selected ion and interpretation of the resulting series.

The first item is the most common source of confusion, and it is arithmetic rather than instrumentation. The distinction between an average mass and a monoisotopic mass is worth reading separately.

What a mass spectrum does not establish

A matching mass narrows the field; it does not close it. Any rearrangement that conserves elemental composition is invisible to the measurement. A D-amino acid substituted for an L-amino acid has an identical mass. Two sequences built from the same residues in a different order have an identical mass. Leucine and isoleucine are indistinguishable by mass alone at 113.08406 Da each. A cyclic form and its linear counterpart differ by one water molecule and nothing else.

This is why mass confirmation is reported alongside a chromatographic purity figure rather than in place of one, and why a purity figure and a net peptide content figure are not the same number.

Where our own position stops

Aurum publishes a mass confirmation and a reverse-phase HPLC purity figure with the wavelength stated. Full fragmentation data are not among the specifications Aurum publishes, so the records we hold confirm that the observed mass agrees with the mass calculated from the stated sequence, and do not independently establish the order of the residues. Ionisation mode is not among the specifications Aurum publishes, matrix or calibration method for every lot, which means a reader cannot recompute our stated mass accuracy from what we make available. Where a question turns on sequence position rather than total mass, our published record does not answer it and we will say so rather than imply otherwise.