You can change the matrix for 1,5 DAN and the fragmentation will occur by In Source Decay. The spectra may however become quite complex as the proteins are not purified. We are working on software and methods to improve that aspect

MALDI-In Source Decay Applied to Mass Spectrometry Imaging: A New Tool for Protein Identification

Anal. Chem., 2010, 82 (10), pp 4036–4045

MALDI In-Source Decay, from Sequencing to Imaging

Top Curr Chem. 2012 Sep 14.

Why do you not try the tryptinization of the whole tissue sample with following MASCOT analysis? If you get unclear results you can use other enzymes for digesten and do later a cross-search in all results. This way I think you can identfify the protein in a not quantitative way and save the downstream processing.

Dear Dennis, I had a similar experience. I tryed to identify a mass peak of about 3800 Da with MALDI-TOF-MS/MS. This was a peptide found in a tissue by MALDI Imaging analysis. I isolated the peptide from tissue exctract by using filter with 30kDa cut-off for the first protein fractionation, and then Pepclean C18 spin column to purify the peptide. MALDI-TOF-MS/MS spectra allowed very clean fragmentation spectra of the peptide. It was a sugar. Did you already check for it? Or maybe, your peptide can be modified (glycosilated i.e.). Did you already check for PTMs on MASCOT search. I would expect LCMS the method of choise to identify your peptide. Good luck!

Yes, the ionisation can quite esily cause fragmentation. Any contemporary MS instrument will have a laser powerful enough. My first choice would be (1) alter the ratios of matix:sample, (2) try different matricies and (3) make sure your sample is concentrated enough, and lastly (4) might try to optimise the LC part prior to the MS.

I agree with the suggestion to try different matrices. Adding 1:1 mol ratio of fructose to a MALDI matrix such as alpha cyano will result in dramatically reduced internal energy of the MALDI-produced ions.

I don't really recall a lot of fragmentation from MALDI...

But I'm not a deep expert in MALDI techniques.

But I can throw some ideas to identify your target:

-you are proposing LC prefractionation, which is a good option, you could also consider (if you have it available) a scan with a SELDI-TOF, it might give you some ideas having a rough first separation, also it works fine with non trypsinized proteins, which is good to see if your target is in any of the fractions there.

-have you considered by any chance a 2D-LC-MS of the digested sample? since you have an ORBITRAP you could proceed like that. Digesting the whole sample (or divide it in fractions before and digest each fraction) with trypsin, then scan them for your target, to see if it's there.

-also you can proceed directly with your proposal, of prefractionation, although I would use a reverse phase more than a SCX, it should retain more proteins and give a better separation.

(also I remember that a SCX column gives a very good profile of separation because it retains few protein species, most of the proteins are negatively charged, so I would expect them to stick more to a SAX), and then analyse the fractions, identify your fraction of interest and scan with the orbitrap for your target.

Using a reverse phase column should allow you also to narrow the fraction, as it could be possible to get a smaller fraction (I'm talking about a gradient elution) and hence get a smaller number of proteins in the fraction to analyse.

Same thing you can get with a 2D-LC-MS method (for example SAX-RP gives a very good range of retained proteins).

it really depends on what's cheaper to run in your lab...

my two cents.

cheers,

Max

Sir, during your sample treatment have you added any peptide cutter? for example like formic acid? You can have a complete list of these on the SwissProt. Moreover, i would agree with you to work downstream to process your sample and if you have no idea of what type of protein your are dealing with, than a reverse phase might help you out in seperation. Best of luck.

Hey eveyrone, thanks for the great suggestions, Edwin the papers were very helpful. Christian, I have considered trypsinization of the sample but the image prep does not give uniform distribution. We need to use a Chip type of spotting to avoid trypsin diffusion and keep the spatial resolution and localization. If the tissue was homogeneous then it would be ok. We don't know if it's peptide or a protein round the 3kDa area. Andor so what you are sayin is that I may be seeing a 3KDa peptide instead of a protein. We also have to get the MS/MS method optimized on the UltraFlextreme and integrate it with the mascot search server. One thing that has been disappointing is that we have not had good tech help from Bruker applicaiton people on this and setting up the correct Lift method in the softwar. They've really have not helped in explaining much at all given the insrument is fairly new. If anyone is using the same instrument I let me know and maybe I can post question for you. It seems that we are definitely underutilizing the marketed potential of this instrument. We've performed LCMS and 1D Page separation to cut out bands at 3KD but do not obtain proteins at that size. We get identified proteins that are much larger. Even looking at the peptides, there are some in the mass range we are looking for but there is a 20Da variation. That's too much for this instrument. From what they say it should be accurate to at least less than 1Da although that depends on the size range you are looking at.

We use the ultaflex II instrument with the smartbeam laser. ISD fragments can be used for T3 sequencing, giving C- and N- ter sequences, depending on the precursor. If you have problems with LIFT, we can help.

Great thanks will definitely let you know.