Maggie,

is this reproducible?

There are certainly some explanations for this. There could be something wrong with your column or with your synthesis.

To investigate the first possibility, you have to repeat your experiment. Your column may not have been packed properly or air was present. If you repeat the experiment after sufficient conditioning of the column and you have the same result, it is probably real.

Then, you could have two isomeric peptides, indicating that something in your synthesis may not have gone the intended way.

Do you have the ability to do MS/MS? That should explain what both products are.

Hope this helps.

Sonja,

No, I have not repeated the synthesis but thanks for your suggestion. I will try that.

Is your peptide a linear peptide? This can be caused by racemization of the peptide during the SPPS. If this is the case the MS/MS should be the same for both of the peaks.

As suggested above run one more HPLC.. Apart from what Dr. S.Hess has suggested, there is another possibility :

check sample injection port. Sample should be injected at one go, if there is an interruption you an get two peaks for same sample

as you observed.

The prinicple of RPHPLC is to separate molecules based on hydrophobicity, but not on molecular mass. Less hydrophobic will have lesser Retention time, while more hydrophobic will have higher RT. You compare the sequence for the aminoacid composition and understand which one hydrophilic compared to other. In mass spec if you do MS-MS you will get the sequence of the peptide, which will identify the peptides appropriately.

What are the solvents you use in the RP-HPLC? Is it water + TFA and acetonitril + TFA?

Did you have cysteine in your peptide-sequence?

Di- or multimerisation of the peptide could be a reason for the two different peaks.

Peptides are separated according to their sequences. The same hydrophobicity, different sequence - different retention time. Most probably, the peptides in original question have residue isoforms, Leu/Ile, or D/isoD. MS/MS will not help, BTW if this is the case (maybe, ECD on FTICR MS).

I would like to add on to the above suggestions. Have you degassed the mobile phase prior to analysis??? Presence of air in solvents might cause air gap during the flow from pump to detector. As a result of air gap your sample might split into 2 half and hence 2 different RT.

It would be helpful if you could provide the following information:

- HPLC system setup and method (others have already pointed out a handful of issues you can trample on while doing chromatographic separations)

- what is the =/= in RT between the two peaks?

- what ms instrument did you used for the MW analysis?

- what is the peptide sequence and how did you perform the synthesis?

- is there any leucine or isoleucine in your peptide?

Hi all,

Thanks to all for some of the answers/advise. Here's an update and what I did.

My synthesis procedure:

This is not a linear peptide. I synthesised the linear sequence on resin (GVAPRSKISPQGY, using Fmoc methods) then exchange Fmoc for Boc, due to instability of Fmoc to base treatments. Then deprotect the two side chains of K (Fmoc-Lys(dde)-OH) and G (Fmoc-Glu(ODmab)-OH) with 2% hydrazine to form a lactam bond (PyBoP/HoAT/DIPEA). Standard TFA cleavage then analysed on RP-HPLC. The first main peak eluted at 14.6 and second eluted at 16.8 min. Looking at peak heights, its almost 1:1 ratio. Then MS using ESI-MS in positive mode and the masses were identical.

What I did next:

I have read up about N-terminal Glu (and also Asp) as been prone to side reactions i.e. Pyroglutmate formation after removal of Fmoc on Glu. So I synthesised the above sequence with the Glu replaced with pyroglutamate (Fmoc SPPS methods). Then ESI-MS the crude and came back with the exact mass as my desired. I then ran that sample down RP-HPLC and overlaid the chromatograms. Peak eluted at 14.6 was the by-product and peak eluted 16.8 min was desired.

Does anyone want to comment? And is overlaying the chromatogram sufficient to say this was the case? Or are there any additional characterisations are needed to confirm?

First: I asked about the presence of L or I AAs because I was thinking about a simple exchange between the two that would leads to isobaric peptides.. but the difference in RT is quite big.

When you say "masses were identical" how many ppm shift are we talking about?

At this point, I think that, as others have already suggested, you have an epimerization/racemization problem.

Time difference due to L/I can be 2 or more minutes. Again it depends on the sequence and the gradient profile (which is not mentioned).

Repeating a question asked by Fabrizio, by how many ppm (i.e. parts per million of molecular weight) do the two compounds differ?

If you have a mass spectrometer that can measure accurate mass then you can almost always tell if two compounds have the same chemical composition. If their compositions differ then the accurate mass of the monoisotopic peak will differ in most cases.

It has already been mentioned that leucine/isoleucine substitution could account for a different retention time. In this case accurate mass measurements will not help.

Another possibility is Gln/Lys substitution. These two residues have the same nominal mass, but the accurate masses of the monoisotopic peaks will differ.

There is also another possibility. Depending on the peptide you are analyzing, there is the possibility that the peptide can exist in differenc conformations. For example, there are at least two forms of hepcidin that have the same accurate mass (hence, the same chemical composition) but different retention times. In the case of hepcidin this is likely due to the presence of different conformers. This is plausible because hepcidin contains several internal disulfide linkages, If these linkages are re-arranged the molecule will exist as different conformers. There have been NMR studies that also confirm the existence of different conformers of hepcidin.

It is also possible for different conformers of peptides to exist without internal disulfide linkages. If the average time to interconvert between the conformers is longer than the LC retention time, and if both conformers can exists under LC conditions, then it is possible to separate them by LC. In such a case there is no difference in the accurate mass.

Another possiblity is to have two compounds with different masses, but an isotopic peak of one may overlap with the peak you are monitoring for the other compound. This would be easily noticed if you are doing full MS scans, but more likely you are doing SIM, in which case you would not normally be able to detect this event.

Most probably you have the same peptide in two different conformations.

Plasmids (circular DNA) for example could be coiled and supercoiled. You can see those conformation by agarose - gel electrophoresis.

May be is a similar situation.

May be they have different structure and conformation or differ in their charge.

Hi Sinclair,

Some quick questions:

How long does it last the cleavage step?

Do you get rid of the TFA afterwards?

What is the final solution composition (the one you leave stand there for 1 hour)?

MW of the peptide?

In your case, could it be that you have an incomplete cleavage? You could easily test this hypothesis by increasing the cleavage time and see if you have differences in the composition.

If you want to test the conformation hypothesis, there are several ways, all depending on the type of peptide you have. For example, if you have cysteines and you think that disulfide bonds are responsible for your conformation change, then you would probably like to treat your peptide with some DTT immediately before injection.

Or you can try to disrupt the AA interactions with salts, buffers or denaturants.

I think you can forget about the incomplete cleavage. Since your final solution does not contain any kind of TFA, no additional or incomplete cleavage.

A test you can try is the following:

Take your peptide and dissolve it using different pHs. I would suggest from 2 to 8 in 1 or 2 pH steps.

This should give you an idea about a pH dependence of the phenomenon you are observing.

Did you check if this peptide can form some kind of cyclization?

To Maggie: What resin did you use? We had recently a lot of troubles with a supplier that sells not enantiomeric pure loaded resin. it results in double peaks in HPLC with L and D aminoacid in C-terminal.

Thanks for all the answers and suggestions. I have made the N-terminal pyroglutamyl peptide (the N-terminal Asn was substituted with a pyroglutamine). Ran that on a MS and it gave exactly the same mass as the desired cyclic peptide. Then I analyse the samples on HPLC, one peak had the exact same elution time as the pyroglutamyl peptide, hence the other peak was what I needed.