I have some questions regarding how to read a MALDI mass spectrum in a paper. I am a biologist by training, so I would like some analytical chemists to help me out.

The researchers ran a western blot, and immunoblotted for a protein of interest. They cut out this band from the western blot, subjected the protein to trypsin digest, and used these peptide fragments for MALDI-TOF analysis. I know for a fact there will be other proteins at the same apparent molecular mass which would have also been included in the analysis, which should be taken as contaminants in the resulting spectrum.

They produced a graph with the absorbance intensity x 1,000 on the dependent axis, and mass:charge (m/z) ratio on the independent axis.

The authors point to two peaks, which they say correspond to two peptide fragments from the tryptic digest of their protein of interest. They provide the supposed peptide fragment sequences and m/z ratio data for these two peaks, and I checked to make sure they are the same as reported. The authors expect to see two peaks: protein of interest is digested to two fragments, A + B.

I have two questions:

  1. Peak 'A' and peak 'B' are of different intensities. Peak 'A' is 20-fold greater than peak 'B'. However, the tryptic digest must yield equimolar amounts of peptide fragments A and B. Is it possible that these intensity peaks can be 20-fold different? Wouldn't the absorbance intensity correlate to total analyzed peptide mass?

  2. Is it normal to multiple absorbance intensity by a (seemingly arbitrary) factor, in this case, 1000? Since the absorbance intensity of peak 'B' is < 0.5 (after scaling up), could this represent noise rather than signal in the analysis?

  • 3
    $\begingroup$ Could you provide a link or a reference to the article in question? I am a bit puzzled: usually identification of a protein based on only two peptides should be considered very carefully in a peptide mass-fingerprint (PMF) analysis, and is probably not valid if many other proteins are present (as peptides with similar masses could originate from other proteins). Did the authors perform any MS/MS experiment to confirm the identification step? $\endgroup$ – PLD Nov 4 '13 at 16:48

Please see the wiki articles for the general method:

Then let's start with question 2: yes, you can and you should. The spectrometer measures in arbitrary units, therefore you can speak only about relative abundances.

And question 1: it depends how easily is the given molecule ionized. So if the fragment A is rich in polar aminoacids, whereas fragment B is mostly hydrophobic and difficult to ionize, than you should expect this big difference in signal intensities.

  • $\begingroup$ Thank you. So if we assume that A is twice as massive as B, and are equally ionizable, then there should be a discrepancy in their relative signal intensities. Shouldn't there be some form of internal control, or is that only if one wishes to quantify absolute peptide masses from the spectrum? $\endgroup$ – user2614 Nov 2 '13 at 23:06
  • $\begingroup$ If the fragments are equally ionziable, then the intensity should be similar. Of course, the m/z will be twice as large for the bigger one. Maybe I do not fully understand your problem. It's m/z which is the important characteristic, whereas intensity tells you little bit about ionizability, ion stability and quite indirectly about abundance. $\endgroup$ – ssavec Nov 3 '13 at 8:51
  • $\begingroup$ Actually you explained the answers to my questions well. I do see the m/z are different by a factor of two. I would assume that ionizability and ion stability would be the same for small peptides, and so I should expect similar band intensities. I think that is enough to be critical of the results I have found in paper. $\endgroup$ – user2614 Nov 3 '13 at 16:22
  • $\begingroup$ @leonardo: There are no real reasons for the ionization efficiencies to be similar for all small peptides. Physico-chemical properties of small peptides can cover a wide range, and you cannot assume that "small peptides" is enough to allow for similar ionization efficiencies. $\endgroup$ – PLD Nov 4 '13 at 16:35
  • 1
    $\begingroup$ In addition, I would like to point out that the intensities in the spectrum are not only related to ionization efficiencies: the entire workflow has to be analyzed, from the initial peptide extraction from the digested band to the mass analysis. For instance, peptide extraction from a gel can lead to strongly different peptide abundances due to pore sizes in the gel if the peptides have very different sizes. (BTW, did they perform the digest directly on the membrane blotted protein or in gel?) A TOF analyzer can also introduce strong variations in intensity when the m/z values are far apart... $\endgroup$ – PLD Nov 4 '13 at 16:40

I know this is an extremely late response, but perhaps this will help another person who is looking at this.

for your first question, the degree of ionization in Maldi depends mostly on the acid-base equilibrium between the matrix substance and the analyte, that is, it's Ka value. And as the above answer has already mentioned, a good way to predict this would be to see how many polar amino acids residues there are. So it is entirely possible that even if the theoretical ratio of two peptides are 1:1, you might obtain a different ratio in the spectrum.

Someone in the comments section has asked whether the authors of the paper in question has done any MS/MS experiments. I happen to have the same question. So far, I have heard of anyone doing peptide sequencing using just MS:it is either done by Edman degradation or by MS/MS experiments (through de novo peptide sequencing). Since you mentioned that they did do sequencing, I'm starting to wonder whether you actually meant a MALDI-TOF/TOF. In that case, fragmentation of individual peptides chosen for the second TOF experiment would depend on the collision induced dissociation parameters. Here is a website detailing the fragment types: https://en.m.wikipedia.org/wiki/De_novo_peptide_sequencing


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy