I am trying to develop a HPLC method to separate tow peptides of 84 and 80 amino acids. I have tried TFA and few RP columns such as Dynamax, Jupitor, etc with no luck. Two peaks are at ~0.5min resolution but can not separate to the basline resolution. The most tricky part is the 4 amino acids are Ala-Lys-Ser-Gln. You can see Lys has + charge so I tried cation exchange method with no luck and the contribution towards the hydrophobicity is also equal so I feel that s why its difficult to separate these two. I would appreicate any of your ideas and suggestions to separate these 2 peptides.
Ananda S.

For the method that gave the best resolution, you could decrease the flow and increase the column temperature. These are the two most common ways of separating peaks that are only partially resolved. Decreasing the injection volume can also help. If you decrease the flow, decreasing the injection volume won't adversely affect your sensitivity that much (if your detector is MS, it will help). You could also try increasing the length of your column, but remember that doubling the length only increases the resolution by the square root of 2, so this can significanly add to your run time without helping your resolution much. If you are using methanol as your mobile phase, you could also try switching to acetonitrile or a 50/50 MeOH/AcN mixture. Column manufactures often can suggest columns or separations if these tips don't help.

This might be a little "out there", but maybe an ion-pairing agent (other than the standard TFA) such as an alkane sulfonate would help separate these with a low pH mobile phase on an RP column. It would be tricky because such IP agents are tricky to use with gradients, and if I'm not mistaken, peptide work on RP supports almost always needs a gradient, but I bet that it can be done with care. Long ago and far away,I looked at a collagen hydrolysate this way after a TFA-MeCN method proved insufficient and it worked reasonably well. Michele's suggestions make sense, too.

Good Luck!

Chris

You did not talk about the column length and the gradient that you are using. If your column is short enough (say 5 cm), you can manipulate resolution by changing the flow rate, keeping the gradient time constant. Or you can change the slope of the gradient by changing the starting composition. There is also nothing magic about the TFA concentration (or pH).
Give us a bit more info on column length, flow rates and the gradient...
There are also some really radical changes that you could make....

Column length is 250mm (X4.6mm). I have mainly used 2 columns Vydec C-18 and Jupiter C18 and tried 0.05%, 0.1% and 0.2% TFA in over a series of gradients. Originally 84 peptide was validated in a method with a 20%-60% gradient of B (80% ACN, 0.2%TFA)/ A DIW/0.2%TFA). I tried several gradients of this sytem and with this coulmn but no luck. My present best resolution was gained using 5%-10% B (50%ACN, 0.1%TFA), 95%-90% A (30%ACN, 0.1%TFA) in 65min using Jupiter C18 (I know its really shallow!!) Still the resoultion is not at the base line and peaks are merged at the base line but at the top of the peak they have ~2 min resolution. This is somewhat strange so i dont know whether the mixed (84vs80)sample I am injecting is bad (got from another lab from their analysis) and there is something hindering the baseline resolution. Thanks for your time and the comments. Flow rate is always 1ml/min and i still didn't think changing the column to a smaller one. I tried some IPR too other than TFA.

Do you really need to have these peaks baseline resolved. Is it really worth the time and effort to resolve these peaks further? Ask yourself these questions and give yourself a honest answer.

Michele, Chris and Uwe: Thank you very much for your comments and suggestions.I am still continuing current method to improve my resolution and I might try with another sample. Chris, I tried HFBA other than TFA and no luck. I will try some other Tetra butyl IPRs if this fails. Michele; I prefer to use ambient temp and I dont even have a column heater and the column length i am using right now is 250mm, but you are right if Vydec or Jupiter have smaller (length) columns it might improve the resolution.

Thanks!

Ananda

Yes, I really would like to resolve these two peaks to the baseline resolution as one of them is an impurity and the method should be validated to detect ~0.1% impurity level.
Thanks!
Ananda

Hi Ananda,

I guess MW of your peptide is about 12.000Da! My recommendation is to go with a material of 100 to max. 150A pore size. Sometimes 300A materials can exhibit a negative effect on efficiency or HETP of your compounds. You also can go to smaller particle size material. Stay with Acetonitril and TFA, and try the 100 to 150A material first. Than you also can play with pH, just inject one sample at pH8 for example and see. Please make your life more easy and forget ionpair agents. With smaller particles than you can optimize your method for all needs, like peak capacity, speed and short cycle time. You also can go than with smaller column ID's, to get a better mass sensitivity. If you need more detailed information you also can contact me by e-mail: g.kratz@tosohbiosep.de

Good luck
Gerhard

Gerhard makes quite valid points, but watch out with reducing column diameter, especially on an older LC with a high void volume - it can be tricky business getting gradients, re-equilibration, flow rates, and injection volumes optimized for a small bore column on such an LC. Otherwise, I concur - worth looking into!

Chris

Hi Ananda,

You should maybe try to use higher homologues of HFBA (perflurinated carboxylic acids). I have investigated a lot the influence of these ion-pairing reagents in the retention of amino acids and small peptides and you will I was quite supriced to notice that while with TFA (and water) for example basic amino acids are eluted in the void volume, with HFBA (and a gradient with ACN)Lys is eluted just after Pro (and nine other protein amino acids while with pentadecafluorooctanoic acid (and a gradient with ACN)Lys and the other basic amino acids are eluted after even the most hydrophobic amino acid tryptophane.

The same goes on with small peptides. So you could maybe try this one (pentadecaflurooctanoic acid-volatile in small concentrations.

Another idea would be to change towards a carbon based stationary phase (Hypercarb for example)if your peptide is not too hydrophobic. In the same way Hypercarb have given totally different selectivities for amino acids and small peptides in comparison with silica based material.

Here are some references for you

J. Chromatogr. A 833 (1999) 147-155.
J. Chromatogr. A 855 (1999) 191-202
J. Chromatogr. A 870 (2000) 245-254

Hope the above helps

Kostas

For the analysis of large molecules, you do not need a long column. I would suggest you use a short, 5 cm column with about the same diameter as the one that you are currently using. Then run the same gradient over the same time period and at the same flow rate. You will get drastically different results. I do not know if they will be better or worse, just different. Now you can change a lot of things, from flow rate to gradient duration to pH or TFA concentration or whatever.

Hi All,

Thank you for very much for valuable comments and suggestion from all of you. I am already trying a column with 100A, 5uM RP column, TFA and ACN gradient now. Next thing is I am trying to find Kostas new IPR. WIll post if I have any luck. My peptide is very hydrophobic.
Thanks again!

Kostas: Already got your papers and it looks very impressive and congrat! on your excellent work. Never used a ELSD detectors before but very interesting work.

Thanks Everybody!

Ananda

I have one comment about adjusting the pH of a mobile phase that contain TFA. I feel it is not worth trying to adjust the pH of a mobile phase that contain TFA because at 0.1% TFA levels the pH of the MP is ~1.9 and if you go to 0.2% its lower ~1.5 i guess and even at 0.05% TFA the pH is 2.1. What do you all think? Now the other question is how much silica of these columns are going to last at these lower pHs? I know a lot of people out there who work wirh Proteins and peptides use these conditions with RP columns. Any comments????
Thanks!
Ananda

The manipulation of the TFA concentration is a tool for making fine adjustments in the chromatography. We have examples where a concentration of 0.1% gives overlapping peaks, while a concentration of 0.05% gives a separation. Of course you are correct: one is changing two things at the same time, the concentration of the TFA and the pH. The stability of the columns depends on the type of column, but we have had good success with TFA using both monofunctionally bonded columns and trifunctionally bonded columns. Of course, you could also use other acids or buffers.

Hi Uwe,
Thanks for your commments and I agree for TFA Mphases its no point adjusintg pH but probably for MP with other buffers its a good option.
Thanks!

Ananda

Hi folks,

From reading the strign of messages you
posted I am wondering whether or not it would
be usefull for Ananda to look into a completely
different eluting system. For example, Anada,
if your peptide (actually a small protein) has
more acidic amino acids (D, E) than basic
ones (R, H, K) it may be worth to try running
your peptide at close to a neutral pH (pH 6.5 in
a buffered system). This way you would
exploit more efficiently the presence of these
acidic AA. Also, since your peptide is very
hydrophobic, using hydrophobic ion pairing
reagents, will only wersen the situation and
render the 4 AA difference of your two peptides
even less apparent. This is what I think
anyway. My approach (after assaying the TFA
system at a low gradient rate and higher flow
rate) would be to use a phosphate buffer at an
appropriate pH for either ion pairing with basic
AA (low pH) or acidic AA (higher pH). I would
suggest using TEAP, in both mobile phases
and at the same concentration, and then try to
vary the ionic strength of the mobile phase.
Remember to keep the ionic strength and pH
constent over the entire gradient. Unless you
wish to vary these parameters during the
gradient (especially ionic strength). As for
stationnary phase, I would definately look at
shorter columns with a good surface coverage
density; this is what is most critical for the
stationary phase. Dont be shy to use C4, C8
or even CN stationary phases.

Remember one thing, you may have to make
compromises as no one system may give you
all the resolution and chromatographic power
you want. Also, as a QC chemist myself,
whatever you come up with as a
chromatographic system, make sure your
sample and sample prep solution will be
stable enough under these conditions. It is no
use of developing a perect chromatographic
system for assaying an impurity if your sample
degrades under these conditions!

Good luck to you and, guys, please comment
on what I have just said. Also, Ananda, keep
us posted of your results.

Regards.

Guy