Chromatography Forum

Might anyone know what is going on?

I am trying to create an LC-MS/MS method to quantify O-acetylpsilocin in plasma. Running several neats has given me a consistent retention time of 3.3 minutes for my analyte, but whenever I perform a simple protein precipitation extraction using a 50 MeOH: 50 ACN mixture, the retention time of my O-acetylpsilocin shifts from 3.3 to 2.6 mins. I have repeated it several times on separate days, and this retention time shift is consistent across these different extractions. I look at 6 transitions for O-acetylpsilocin, and they are as followed:

247 -> 58 (My quantifier transition)
247 -> 160
247 -> 89
247 -> 77
247 -> 63
247 -> 115

BUT here's the funny thing: it seems like this retention time shift in plasma only affects my 5 qualifier transitions. These 6 transitions appear at the correct 3.3 mins mark in my neats though. I have tried using other precipitation mixtures, but I am also analyzing other drugs in my assay, and this 50 MeOH: 50 ACN precipitation mixture enables me to get the best recoveries across all my drugs. But WHAT IS GOING ON WITH O-ACETYLPSILOCIN!!!

Do you guys have any suggestions as to what I can do to troubleshoot this issue?

If the retention time shift affects not all mass transitions in the same way, you have isomers which are separated by some matrix effect. I don't know about the drug or it's fragmentation pattern, but it is impossible to have two different retention times for different daughters of the same parent molecule, because fragmentation happens after the chromatographic separation.

Hi there,

I may be a little confused by some of the wording but when you say transition, this is just another way of saying the fragmentation patterns for your molecule? As in your base peak is the 247 while the fragments (from the MS) are to the right?

If that's the case, and all of the fragments are different post plasma derivatization, could it be the case that your derivatization or plasma is causing a chemical change of the molecule? This would result in some completely new fragments and potentially some identical ones.

I just looked up the structure and it has quite a few functional groups for such a small molecule, and it looks like it can also be protonated/deprotonated. What does your mobile phase look like? Additionally, what is the solvent used for injection of the extracts vs the standards? Could it be as simple as a proton? haha

Have you used your neat samples in a spiked plasma test? I suppose I'm not sure if these are coming from actual tissues or are just always spiked into the solution.

Anyway, get us some more information and hopefully we'll get to the bottom of this issue!

You've provided no details about your chromatography, so I'm free to speculate about a scenario. The dimethylamine- group in acetylpsilocin's side chain is in fact ionizable. That means that it's going to have a counterion at any pH low enough for silica not to dissolve. Coming straight from plasma, it's reasonable to assume that the counterion for the amine is Cl-. Your precipitation step will get rid of much/most of the NaCl from plasma. Now, if you proceed to run the analysis on a reversed phase column with TFA in the mobile phase, then the amine group's counterion will be trifluoroacetate. That's more hydrophobic than chloride ion, and would account for the increase in retention time for the resulting ion pair. You can assess this by using a chloride salt in the mobile phase instead of TFA.

If this speculation is off-base, then kindly provide some details about your chromatographic conditions.

Thank you so much for all the replies thus far! I appreciate all the help I can get really, but here's additional information about my LC-MS/MS method.

@AndyAlpert and @TylerSmith123 If I am missing any information please let me know! This is my first time posting on this forum haha.

@TylerSmith123 My apologies if I may be using confusing terminology. To be clear (hopefully!), I meant that my [M+H]+ ion is m/z 247, while my product ions (after fragmentation) are m/z 58, 160, 89, 77, 63, 115. I am not performing any derivitization during my sample prep. I share a similar hypothesis as you do, in that I was thinking whenever I spiked my drug in plasma, there might be some interference in the plasma that may be reacting with the analyte. But if this was so, I am unsure as to how I can overcome this issue.

You've provided no details about your chromatography, so I'm free to speculate about a scenario. The dimethylamine- group in acetylpsilocin's side chain is in fact ionizable. That means that it's going to have a counterion at any pH low enough for silica not to dissolve. Coming straight from plasma, it's reasonable to assume that the counterion for the amine is Cl-. Your precipitation step will get rid of much/most of the NaCl from plasma. Now, if you proceed to run the analysis on a reversed phase column with TFA in the mobile phase, then the amine group's counterion will be trifluoroacetate. That's more hydrophobic than chloride ion, and would account for the increase in retention time for the resulting ion pair. You can assess this by using a chloride salt in the mobile phase instead of TFA.

If this speculation is off-base, then kindly provide some details about your chromatographic conditions.

Hi there,

I may be a little confused by some of the wording but when you say transition, this is just another way of saying the fragmentation patterns for your molecule? As in your base peak is the 247 while the fragments (from the MS) are to the right?

If that's the case, and all of the fragments are different post plasma derivatization, could it be the case that your derivatization or plasma is causing a chemical change of the molecule? This would result in some completely new fragments and potentially some identical ones.

I just looked up the structure and it has quite a few functional groups for such a small molecule, and it looks like it can also be protonated/deprotonated. What does your mobile phase look like? Additionally, what is the solvent used for injection of the extracts vs the standards? Could it be as simple as a proton? haha

Have you used your neat samples in a spiked plasma test? I suppose I'm not sure if these are coming from actual tissues or are just always spiked into the solution.

Anyway, get us some more information and hopefully we'll get to the bottom of this issue!

Hi Sam,

We're happy to have your post! No need for apologies.

As for your sample, it seems like the process for making your neats and spiked samples are nearly identical, but with one you are getting the shifted RT peak, while in the other (plasma) you are not. I may be a little confused, but I believe you also said that the fragments are affected when you compare the o-acetylpsyilocin neat sample to that of the plasma-extracted sample. When you prepare the plasma sample for extraction, is there a set amount of time you wait before extracting the sample? I'm certainly no expert, but some enzymes or something within the mixture may be acting on the compound in the time prior to extraction. Perhaps spike the solutions extractant (although I know this could not be implemented normally) following extraction with o-acetylpsyilocin to make sure that there are no enzymes affecting this compound (you'll probably see 100% recovery but whatever, it's just a test afterall). Other than that, the only thing that I can see different is using the acetonitrile in one extraction versus the other which should make no difference on the molecule. The one thing I could think is that the acetylpsyilocin is being converted into psilocin but you have the standard for that anyway, but the deuteration may affect the retention of the standard peak (I'm not an expert on this so take that suggestion with a big grain of salt). Anyway, I hope someone else can shine some light on this! All the details you have added will definitely open up the conversation to more ideas.

Tyler