Mass spectrometry: identity, not purity

HPLC tells you how much of what it can see sits in the main peak. It does not tell you what the main peak is. That question belongs to mass spectrometry, or MS, which measures the mass of molecules. For a synthetic peptide the central check is simple: does the measured mass match the mass calculated from the intended sequence?

How a mass spectrometer works

Every mass spectrometer does three things: it gives molecules an electric charge, sorts the charged molecules (ions) by mass, and counts them.

• Ionization. Only charged particles can be steered by the electric or magnetic fields inside the instrument. For peptides the two common methods are electrospray ionization (ESI) and matrix-assisted laser desorption/ionization (MALDI). In ESI, the liquid carrying the sample is sprayed from a fine capillary held at high voltage; the charged droplets shrink until charged molecules escape into the gas phase. In MALDI, a laser pulse lifts molecules out of a solid matrix. ESI is gentle, so peptides reach the analyzer intact, usually carrying extra protons. It can be connected directly to an HPLC, which is called LC-MS.
• Mass analysis. The analyzer sorts ions by their mass-to-charge ratio, written m/z: the ion’s mass in daltons (Da) divided by the number of charges it carries.
• Detection. The detector counts the ions arriving at each m/z. The output, the spectrum, plots ion intensity against m/z.

Why one peptide gives several peaks

In electrospray, a peptide usually picks up more than one proton. Each proton adds one positive charge and 1.00728 Da of mass. An ion carrying z extra protons therefore appears at

m/z = (M + z × 1.00728) ÷ z

where M is the neutral mass of the molecule. Take an invented peptide, Peptide A, whose neutral monoisotopic mass is 1,800.00 Da:

• with 2 protons: (1,800.00 + 2.015) ÷ 2 = 901.007
• with 3 protons: (1,800.00 + 3.022) ÷ 3 = 601.007
• with 4 protons: (1,800.00 + 4.029) ÷ 4 = 451.007

So one molecule shows up as a family of peaks, called the charge-state envelope, and none of them sits at 1,800. This is why a raw electrospray spectrum can look puzzling at first glance.

Deconvolution: from several peaks back to one mass

Two neighboring peaks in the envelope are enough to work out both the charge and the mass. If a peak at m/z m1 carries z charges, and its neighbor at the lower value m2 carries z + 1, then

z = (m2 − 1.00728) ÷ (m1 − m2), and M = z × (m1 − 1.00728)

In the invented example the lab measured peaks at m/z 901.018 and 601.013. The charge is (601.013 − 1.00728) ÷ (901.018 − 601.013) = 600.006 ÷ 300.005 = 2.00, and the mass is 2 × (901.018 − 1.00728) = 1,800.02 Da. Software does the same for every peak in the envelope and reports a single, deconvoluted mass. The lab then compares it with the expected mass calculated from the sequence: 1,800.02 Da found against 1,800.00 Da expected, a difference of 0.02 Da.

There is a second way to read the charge. Zoom into any one peak and it turns out to be a small cluster, because about 1.07% of the carbon found in nature is the heavier isotope carbon-13, which adds 1.00335 Da. Inside a cluster the peaks are 1.00335 ÷ z apart: about 0.50 for a 2+ ion and about 0.33 for a 3+ ion.

That cluster also explains why a certificate should say which mass it reports. The monoisotopic mass counts every atom as its most abundant isotope. The average mass uses each element’s average atomic mass, which reflects the proportions of its isotopes found in nature. For a peptide of around 1,800 Da the two differ by about one dalton, so a comparison is only fair if the expected and found values are the same kind of mass.

What a mass match proves, and what it does not

A measured mass that agrees with the calculated one is strong evidence that the main component is the intended peptide. That is what identity means on a certificate. It is not evidence of purity, for three reasons.

• Peak size is not amount. Different molecules turn into ions with very different efficiency, so peak heights for different species are not proportional to how much of each is present. A strong signal at the right mass can sit alongside a large amount of an impurity that ionizes poorly.
• Same mass, different molecule. A mass check cannot see impurities with exactly the intended mass: a chain with two residues swapped, a residue flipped into its mirror-image form, or leucine where isoleucine should be, since those two amino acids have identical masses.
• The rest of the powder. Water, counterions and salts are not part of the peptide’s mass, so a mass match says nothing about them; see Purity, net content, net peptide content and salt forms.

A spectrum can still flag some impurities. A form 16 Da heavier (15.995 Da) usually means one added oxygen atom, an oxidized form. A mass one residue lighter than expected points to a deletion sequence, a chain missing one amino acid (see How peptides are made). A mass about 22 Da above the expected one (22 divided by the charge, in m/z) is usually a sodium adduct: the same molecule carrying a sodium ion picked up during ionization, rather than a different molecule in the sample. Tandem mass spectrometry (MS/MS) goes further: it breaks the ion into fragments and reads the sequence piece by piece, which is stronger identity evidence than an intact mass alone.

Reading the MS line in a lab report

• The method is named: ESI-MS, MALDI-TOF or LC-MS.
• The expected mass is stated, and it says whether it is monoisotopic or average.
• The found mass is stated, with the acceptance window the lab applied.
• The spectrum is attached, showing the peaks the mass was calculated from.

Then read it next to the HPLC line. Identity by MS and purity by HPLC answer different questions, and a certificate needs both; see How to read a certificate of analysis.