Chromatography Forum

Hello all. I am developing a method to analyze peptides with molecular weights in the 3000 - 4000 dalton range, in a biological fluid by LC-MS/MS. I have determined some useful MRM transitions using synthetic standards of the peptides, but have hit a stumbling block with the LC. I've tried a few different columns, pH modifiers, and tried ACN versus MeOH as organic modifier. I get good chromatography from a Luna C18(2), 2.1x50mm, 3µm, 100 angstrom pore size column, with water-ACN gradient, and 0.1% formic acid as pH modifier. The problem is with carryover, and something that looks like adsorption.

When I run a solvent based calibration curve, I see carryover as high as 10% (depending on the peptide) with the blank following the high standard. A non-injection blank after the high standard indicates that some carryover appears to be coming from the column, although I suppose I can't rule out active sites elsewhere in the flow path. A flow-injection experiment with standards and blanks indicates that some carryover is coming from the autosampler. My needle-wash solvent is presently 1:1 MeOH/water, which works well with some small molecules but might not be ideal here. Any suggestions for wash solvents or sample additives that will help minimize carryover will be appreciated. I have already tried adding bovine serum albumin (at 10x the molar concentration of my high standard), and it actually seems to worsen the carryover.

Also, any suggestion as to a column that won't hang on to my compounds through to the next analysis, will be appreciated. So far, all the columns I have tried have been silica based, C8 or C18. For mobile phase modifiers, I'm somewhat limited to what gives good MS sensitivity. I've already ruled-out acetic acid, formic acid/ammonium formate, and high pH (ammonia) for these peptides.

Another bad thing I have noticed: the signal seems to drop of rapidly when comparing the high standard to lower standards (or increases rapidly from low to high, depending on your point of view). In other words, let's say I get 100,000 area counts with my high standard. I expect to have something close to 10,000 area counts at 1/10 the concentration, and 1,000 area counts at 1/100 the concentration, but actually I might see 2,000 counts at 1/10 the concentration, and not detected at 1/100! I wonder if this could be caused by adsorption, and if so, could it be related to the carryover described above. I am using plastic (no glass) for all standard preparation and storage.

If you can't tell, analysis of biomolecules is new to me. I don't recall ever having this much difficulty with small molecules!

Remember that to get carryover you have to have some mechanical action, like the actuation of an injector, needle, etc., or a disturbance in mobile phase equilibrium. If something adheres somewhere, without such perturbations, you either see nothing (analyte is partly, but completely stuck somewhere) or you get a rise in baseline (very strongly retained, moving out slowly....). So on first thought you could have some peptide stuck in a part which is not swept by mobile phase until the next injection, or you are disturbing material which has gotten stuck to the stat. phase, frit (areas constantly swept by mobile phase) with a subsequent injection that differs from the mobile phase.
If the latter is the case (peptide stuck to permanently swept parts) you have to play with your mobile phase, like using a buffer, changing ionic strength (chaotropy, etc.), maybe trying Propanol or iso-propanol at diff. %, or detergents.

Thanks for the suggestions, HW. Regarding the peptide that appears to be sticking to permanently swept parts: from the experiments I have done so far with changing the mobile phase pH, I am worried that I will lose sensitivity if I move away from the low pH 0.1% formic acid that I am using. I will probably give the isopropanol a try, as I don't expect it to hurt sensitivity. Any ideas on the percentage of isopropanol that would be useful? (I know if it's too high, pressure becomes a problem).

Anyone know if polymeric columns are useful in cases like this? What about these new silica-hybrid type packings?

I don't have any personal experience with peptides.


However, i went to a discussion group one evening at ASMS in San Antonio. One of the presenters was talking about solving carryover problems using a monolithic column.

Sorry, can't remember person's name..

I did a google search using "monolithic column for peptides" and found some interesting results.

MG,

We had some carry over problems (mainly with plasma samples) which we solved. These problems were mainly due to the autosampler and then a problem that has to do with the use of capillary columns the way we use them (you do not use capillary columns so you do not have a problem there).

The problem from the autosampler was solved with the use of different washing protocols that cleans all the parts of the autosampler. The washing solvents are mainly water and ACN but I can look at the exact composition and communicate them to you.

About your non-linear response, I agree that it looks like somekind of adsorption, probably having sites that absorb your peptides of interest. Are these very chelating peptides (i.e. phosphopeptides or similar)? In general, each time we have a new column, before using it for our proteomics samples we run some Qc standards which is actually a mixture of peptides and proteins as described at: Purvine et al. Omics-a Journal of Integrative Biology, 2004 8 79-92.

We use mainly Phenomenex Jupiter as our stationary phase, TFA/acetic acid as mobile phase additive. The choice of the stationary phase was done after we tested several other, although that study starts to get old (for example we got better results when we recently did some preliminary studies with the Waters Aquity stationary phase). We do use some other stationary phases and instrumentation in general when dealing with chelating peptides...

Hope the above helps a little bit...

PS: If you think that you are having problems with the peptide separations wait until you get to the intact protein world

Sorry, MG, I didn´t register the MS yesterday. These additive limits are the real problem with hyphenated MS. If MS compatibles don´t do the job it may be best to include another step prior to MS?
The propanols are not starkly different from other organic modifyers, regarding method development, how are they in MS?
Also, you ought to know how others have handled these peptides (for instance if you have standards, someone knows how to handle them), you may find something useful there. If you removed your adsorption problem, but still get carryover you may gert lucky, like Kostas, with washing steps (discussed a long time ago: Sometimes carryover can only get licked by taking some injection valves apart).
You might have seen other discussions, and know that my suggestion regarding TFA is to use it only as a last resort (I am an old biased sack, Kostas?).

HW,

Indeed, there is no problem of using propanol with the MS. As of TFA, you might experience some sensitivity loss. I would give it a try though, we saw that in addition with acetic acid in the mobile phase things are better, your mobile phase pH will also be lower.

Anyway, I do not think that the TFA will have any influence in the carry over effect (except if the pH has to do something with the active sites etc...).

As for Hans if not biased, I would put it more gently, something like: strongly opinionated against the TFA (does it sound better Hans?).

Generaly speaking though, most of the people in the proteomics field are using 0.1% formic acid and not TFA, mainly because of the sensitivity loss, but they sucrifice a little bit of their peak capacity, which is important for us that we are doing global proteomics...

Finally, we are using some high pressure valves from Valco which have somewhat different design than other valves. Having said that, we saw that is even easier to handle with carry over in our commercial capillary LC from Agilent (low pressure system).

Finally Hans, I am feeling like buying a ticket and going to Las Vegas for the weekend. I was lucky with our carry over problems, so who knows???

Thanks Kostas.

Well that choice is a bit like deciding between beelzebub or the devil...

I seems that two areas could definatly stand some further development, especially in the discussed context: Injection valves and the coupling to MS. Some people appear to work on handling non-volatiles (phosphate buffers, etc.) regarding the latter?