IEX or HILIC for peptides - any experience?

[aschultze]

We are dealing with a peptide of ~1900Da, which contains a total of 13 amino acids. 4 of those are basic AA's (3 Arg, 1 His), and 1 is an acidic AA (Asp). We have developed a RP-HPLC method for purity evaluation which works quite well. However, it is always desirable to have an orthogonal method. For this orthogonal method, I am considering an ion-exchange or maybe a hydrophobic interaction methodology. Has anyone have experience with either methodology for peptides and can share their findings? I am particularly interested in potential columns and mobile phase systems.

Thanks in advance.

[Uwe Neue]

We did HILIC for peptides a few years ago. The focus was on peptides that were unretained on a C18, however, the same method might do well for peptides in general.

Column: Atlantis HILIC silica

Gradient: A: 90% acetonitrile, 6.5 mM ammoniumacetate pH 5.5, B: 6.5 mM ammoniumacetate pH 5.5; Gradient from 90% A to 60% A.

If nothing else, this is a good start.

I think HILIC will do better than HIC.

[aschultze]

Uwe, Thanks for the feedback. I'll try those conditions but will likely use a newer column (XBridge HILIC). I'll post my findings.

[Kostas Petritis]

Another orthogonal way would be to use a SCX (we currently use Polysulfoethyl aspartamide) and use mobile phase containing 25% ACN and a gradient from 10 mM of salt at pH 3 (but you can use a higher pH as your peptide is rather basic) to 500 mM. Use ammonium formate if you want volatile buffers otherwise there are plenty of other alternatives if you do not care about mobile phase volatility...

[aschultze]

Kostas, thanks to you as well for your feedback. SCX was the other methodology I considered, so I'll give that a shot as well.

[aschultze]

We did HILIC for peptides a few years ago. The focus was on peptides that were unretained on a C18, however, the same method might do well for peptides in general.

Column: Atlantis HILIC silica

Gradient: A: 90% acetonitrile, 6.5 mM ammoniumacetate pH 5.5, B: 6.5 mM ammoniumacetate pH 5.5; Gradient from 90% A to 60% A.

If nothing else, this is a good start.

I think HILIC will do better than HIC.

I've done some preliminary experiments with a Waters HILIC column. So far, I can't retain my compound on the column at all, even when starting at 100%A. I used 10mM acetate pH 5.5 in A and B. I also deal with tremendous backpressure issues at high levels of A, so I am running at a pretty low flowrate to not exceed the pressure limit. So all in all, I had not much success with HILIC yet, trying SCX next as suggested by Kostas.

[Einar Ponten]

It is not good idea to use 100% acetonitrile as starting mobile phase.
1. You always need to have some water in the mobile phase to establish a water layer on the stationary phase. A basis for the retention mechanism.
2. At 100%ACN it is a considerable risk that you get precipitation of these analytes. The high back pressure indicates this problem.

We provide a 2D-LC kit containing both a RP and a ZIC-HILIC column as well as three (x2) freeze dried tryptic digest of some common proteins. By using this kit it is easy to calibrate your 2 dimensional approach and get started.

Clearly, RP and HILIC are orthogonal and suitable for your purpose.

[Kostas Petritis]

Einar,

Aschultze said that he/she uses 100% A and not 100% ACN. Mobile phase A consists of 90% ACN as suggested by Uwe...

[Einar Ponten]

Thanks for the correction! Starting at 90% ACN should be a proper procedure. Anyway, something is confusing here.

Uisng 90% ACN on a plain silica column should give very low back pressure, typically for HILIC. Here the opposite is reported?

I suggest that the column is carefully washed and solubility in 90% ACN is tested off-line before next injection. Why start at 90%ACN for this seemingly basic peptide. It is probably sufficient to run between 80-->60% ACN

[danko]

I bet the pressure wasn’t high prior to the first couple of injections.
It’s plain (to me anyway) that the peptide precipitates upon contact with the mobile phase.
Actually I know many peptides that will precipitate even at lower ACN concentration than 90%.
The situation will be worsened greatly if the mobile phase pH is approx. the same as the pI of the peptide/protein.
Given all these probabilities of failure, I would clearly select the ion exchange option.

Best Regards

[Uwe Neue]

I agree: the high backpressure indicates precipitation. You should have seen some peaks though somewhere, or the peptide did not elute at all.

[goxy43]

I am doing HILIC of peptides and did try several columns. We all have to bear in mind that HILIC is not HILIC on all "HILIC" columns. Basically, I want to say that the retention on one particular HILCI column might strongly differ from retention on another column. I did use the columns from SeQuant, Phenomenex and TOSOH and they all behave in a different way under identical chromatographic conditions.
Why not simply try to use 80% AcN + 0.1% formic acid as A and 5% AcN + 0.1% FA as B? For me, that mobile phase was the one that showed the best performance with a broad range of both synthetic and natural peptides from tryptic digests.
Another approach: I would try the PolyWAX LP with ammoinum acetate - 20 mM pH =2.7, 70% AcN as A + 200 mM TEAP, pH 2.7 + 10% AcN as B. The separation is perfect!
Success

[Mattias]

Going back to the original question: I am a bit confused by the word othogonal in terms of switching between RP and HILIC.

I know SeQuant claims that HILIC is an orthogonal method to reversed-phase their material (over and over). For me an orthogonal method is a method that gives a separation that is totally independant on the first method. When you compare HILIC and reversed-phase chromatograms, they often look similar in elution order - with the difference that the elution order is backwards.

I believe that IEX or SEC are more othongonal to RP than HILIC.

[HW Mueller]

Mattias, we have just been through this:

http://www.sepsci.com/chromforum/viewto ... sc&start=0

What good does a strict definition of orthogonality do? What good does it do to "confirm" a RP analysis with IX if the latter has all the peaks piled up one on the other (a good possibility).

[Mattias]

OK, I missed that one! You seem to have been through all aspects of this already.

As an engineer with lots of mathematics behind me, I will still refuse to use the word orthogonal. But I can still use HILIC as a complementary method...