Phospholipid matrix effects

[Anna01]

We have problems with phospholipid matrix effects causing suppression of one analyte (non-polar). I am working in ESI mode, ACPI is not an option.

We use cation ion-exchange to cleanup the matrix, which is rather successfull with the exception of one (probably) phospholipid substance.

When doing MRM on 104/104 and 184/184 (common phospholipid transisitons) the 184/184 is comparable to water. In the 104/104 transition there is a variable, but often present coeluting peak. Experiments confirm that, when present, this causes matrix effects up to about 20%.

I have tried different gradients and temperatures. Modifying pH from 2-3.5 doesn't seem to make any difference. Neither doing switching from MeOH to ACN (actually much worse). I can not separate my substance from the interfering peak.

Cleanup method
Load sample on cation exchange sorbent (Waters MCX)
Wash with H2O
Wash with 20% MeOH (Acidied or basiedfied)
Elute with ACN
Dilute with 1:2 water

HPLC
Gradient 45 - 95% MeOH 0.1% FoAc in t=0-3

How could I improve cleanup? Suggestion on HPLC parameters to change?

[Uwe Neue]

Will your analyte elute with methanol instead of acetonitrile?

(I assume that your analyte is not ionic or ionizable. Is this correct?)

[Anna01]

Yes. Initially I eluted with MeOH, but inspecting the 184/184 transitions I found that ACN to give slightly cleaner extracts.

Here are some data from the elution experiments:
MeOH and ACN refers to elution with 100% MeOH or 100% ACN
The percent values is the AUC of the substance divided by the same amount spiked in H2O.

MeOH ACN
Tissue 1 90 % 100%
Tissue 2 78 % 80%
Tissue 3 71 % 80%


I am doing analysis of tissues, very rich in phospholipids I believe.

Water's documentation for MCX advice to wash with H2O with formic acid, than elute with MeOH. I guess this is rather general. One aim is probably to preserve the ability to elute basic substances in another elution step. I am not interested in basic substances, just neutral non-polar fat soluble.

I have done breakthrough studies. All substances elute at about 25% MeOH adjusted to acidic, basic or neutral pH.

[Uwe Neue]

Are you familiar with our publication in J. Chrom. B 852? Also available at http://www.waters.com/webassets/cms/lib ... 2135en.pdf.

It contains a few other ideas that might be useful.

[Anna01]

Yes, i am familiar with this. This paper also suggests to elute with ACN.

[Uwe Neue]

I would first try to bind the phospholipids by adding a strong acid (HCl) to your eluting solvent. You can try both methanol and acetonitrile and see which one will work better.

[Anna01]

I will try eluting with MeOH and/or ACN with 1% HCl.


I don't know if I fully understand how phospholipid interacts with the MCX sorbent. Sure, there is a reverse-phase mechanism and there is a quaternary amine group that might interact through cation exchange. But there is also a phosphate group that confuses me.

From my experiments it looks like basified washing steps and basified elution releases more phospholipids. Would it make sense to include 20% MeOH with 5% NH4OH as a washing step? Or H2O with NH4OH?

[Uwe Neue]

I will ask tomorrow.

You did not tell us if and how you acidify the sample to start with, i.e. before the loading step.

[ErinChambers]

I have 2 thoughts: have you tried reducing the % of organic in your final SPE elution? You mention that in breakthrough studies, the compound begins to elute with 25% organic (if I have understood correctly), what is the minimum organic % required to elute the maximum % compound? If possible, I would definitely suggest reducing the organic content of the elution if possible. The other thought I have is to try changing to a different mixed-mode sorbent, to encourage alternate selectivity. Are there any ionizable functions on your compound? It appears to bind by reversed-phase to MCX, is it possible it could bind by ion exchange to another sorbent such as MAX or WAX (strong and weak anion exchangers)? This would potentially provide a very selective separation between the compound and phospholipids. Alternatively, binding the compound to WCX (weak cation exchange) by one mechanism or the other could provide different results and perhaps the selectivity you are looking for.

[Anna01]

To Uwe:
I acidify the samples by diluting them 1:1 with 4% H3PO4 in H2O. I am running samples eluted in 0.1N MeOH and 0.1 ACN now, like you suggested.

To ErinChambers:
Yes. I have tried reducing the organic content of the final SPE elution. Going down as low as 60% MeOH gives recovery of about 80-90%. Unfortunately, eluting with anything else than pure MeOH or ACN results in major phospholipid contaminants in the eluat. In fact, such eluate give as much as about 60% phospholipids compared to ACN precipitation. My guess is that this is because MeOH and H2O has protic properties, and thus partially "unlocks" the ion-exchange. Am I right?
My analytes are neutral and not ionizable. I doubt it would bind to any other ion exchange sorbent. It binds to MCX by reverse-phase alone.

Neutrals are a pain to work with when confronted with phospholipids, because they tend to coelute and reverse-phase upconcentration just fortifies the problem.

[Uwe Neue]

If the acidified elution solvents give worse (instead of better) results, we try the opposite direction and try to elute your analyte with acetonitrile/methylene chloride.

[ErinChambers]

You are right about the use of a more protic solvent versus an aprotic solvent such as ACN for elution- 100% ACN will typically give you a cleaner elution if analyte solubility is okay. In addition, phospholipids are more soluble in methanol than acetonitrile, so you will solubilize more of them when MeOH is used.
I understand that your analytes are neutral and not ionizable; however, they will bind to the reversed-phase portion of any of the other mixed-mode sorbents I mentioned, often with desirable changes in selectivity. It might be worth screening some of the other sorbents to see if the reversed-phase elution is any cleaner, indicating that the interferences may have bound by ion-exchange.
On the chromatography side, one thing that has helped me separate an endogenous interference from my steroidal compound in the past when all else failed was the use of a 1mm ID column. If you have access to a 1mm column, it would absolutely be worth trying. In my case, I was trying to analyze a steroid/hormone and was unable to separate it from a co-eluting, isobaric interference by any means other than the 1mm ID column.

[Anna01]

I have run the samples (6 different samples + substance added to H2O, eluted with acidified ACN, MeOH or neutral ACN). Acidifying the eluate solvent did not help to improve matrix effects. Overall suppression increases from about 15-20% to 30%, compared to using neutral ACN.

Monitoring phospholipids by MRM 184/184 and 104/104 also shows (much) higher levels of phospholipids.

[HW Mueller]

This dicussion is difficult to follow.
Anna01, are you trying to get good MS or are you trying to find the correct amounts of phospholipids in unknowns? It seems that if the latter is the case you should work on the chromatography to be able to introduce cleaner samples to the MS? For instance a two or even three step (dimensional) chromatography?

[Anna01]

I am trying to measure neutral steroid substances in tissues, but large amount of phospholipids prevents reliable MS/MS quantification due to ion suppression. Even using an deuterated analogue can not fully adjust for the matrix effects.

Uwe: You mean eluting with ACN:Methylene chloride? Any thoughts on what percentage of ACN?

Does anyone have further suggestions?