diagnostic ions

[pstaunton]

Hi.

I was curious to know how unique precursor to product ion transitions are selected for an analyte. May seem like a very simple question. not sure

Thanks

[Alexandre]

If there is mol. ion then it is easy (for regulatory, forensics purposes mol. ion has more convincing power) if not, then it depends, give more info. Adducts are not advisable either.

[pstaunton]

lets say we take the cocaine metabolite cocaethylene which has a molecular mass of 317 which would correspond to the molecular ion, im assuming we take the two most abundant ions after that as the daughter fragments. How specific for cocaethylene would this be? im assuming that the fragmentation pattern will stay the same as long as all mass spec parameters are held constant. Basically they will show up scan after scan.

[Jetjamnong]

lets say we take the cocaine metabolite cocaethylene which has a molecular mass of 317 which would correspond to the molecular ion, im assuming we take the two most abundant ions after that as the daughter fragments. How specific for cocaethylene would this be? im assuming that the fragmentation pattern will stay the same as long as all mass spec parameters are held constant. Basically they will show up scan after scan.

pstaunton
I use the following ions for Cocaethylene analysis.
196.13 (100%) as quantify ion
318.17 (6%) and
150.09 (12%)
with Collision energy 30 eV (ESI-LC-MS/MS)

(Cocaethylene, C18H23NO4 MW = 317.162160; M+H+ = 318.169985)

Jetjamnong

[lmh]

To answer your question about specificity (using Jetjamnong's ions as an example)
If you monitor the transition 318>196 you are looking at a loss of 122, which in this case is almost certainly benzoic acid. Where something has a benzoate group, it's a common loss, so this transition is "fairly" specific for benzoate compounds. In addition you are obviously only going to see things that have a parent mass of 318. Further, they will have to elute at the right time (your chromatography adds selectivity).

There are undoubtedly other chemicals (somewhere in the universe, if you look hard enough!) that will do exactly the same thing. For example, very similar isomers will often nearly-coelute and fragment identically, but are they likely to be present in the sample? Selectivity must be seen in context: a method is good enough if it can differentiate the target chemical from all the other things it will ever see.

Jetjamnong's method has further selectivity by looking at the other ions. When you look at 318 you are looking at how "tough" the parent molecular ion is: at a given energy, some molecules will disappear completely, others will fail to fragment. 150 is probably loss of benzoate and the ethyl group as ethanol (I'm guessing) so it adds further selectivity of looking for another group that must be present.

A "bad" MRM method will monitor a transition such as loss of 18. This is almost certainly water from a hydroxyl group, and since vast numbers of biological molecules have hydroxyls, it's not much more selective than SIM looking for the parent ion. It's a bit better, and will remove noise, but that's about all. Nevertheless there are plenty of methods around that look for non-specific losses like this.

So my personal answer is: unless someone is really setting out to give you a very difficult sample, Jetjamnong's method is about as selective as you can get, but there is always a possiblity of being hoodwinked, and we should always keep that in mind.