Ways to prevent neutral loss on LCMS

Discussions about GC-MS, LC-MS, LC-FTIR, and other "coupled" analytical techniques.

14 posts Page 1 of 1
Hey guys,

My colleagues and I are trying to elucidate the CYP metabolized product of a small molecule (non-peptide), but we are experiencing some difficulty since the molecule quickly looses 1 or more H2Os as neutral loss and then fragments. However, it does so in such a way that we are unable to identify the fragments since the neutral losses are adding to the complexity of this problem.

Are there any ways to suppress or stop the neutral loss of water before the molecule is fragmented, for example, different collision energies or different instrument set ups? Perhaps are there ways of derivatizing the -OH groups to prevent the C-O bond from breaking so easily? Any insight would be greatly appreciated.

Cheers guys!

Clement
I had a similar problem about 15 yrs ago, while using an API source.
At normal operating temps the molecule readily eliminated an alcohol. Reducing probe and source temps lowered the energy content of the molecule and enabled me to determine the true [M+H]+.
However, there was significant carry-over from injections because of condensation and deposition of the sample in the source etc. Readily removed by raising temps again.

If you are using ESI, then suggest you work with an API source as above.

Regards
JMB wrote:
I had a similar problem about 15 yrs ago, while using an API source.
At normal operating temps the molecule readily eliminated an alcohol. Reducing probe and source temps lowered the energy content of the molecule and enabled me to determine the true [M+H]+.
However, there was significant carry-over from injections because of condensation and deposition of the sample in the source etc. Readily removed by raising temps again.

If you are using ESI, then suggest you work with an API source as above.

Regards


Thanks for the quick reply JMB. I can try lowering the source temp and robe temp since we are not doing long sequences. Carry over shouldn't be a problem.

The reason why we are trying to prevent neutral loss before fragmentation is because there are 2 -OH groups and 1 -NH2 group on the molecule. If the CYP oxidized site eliminates itself through neutral loss amongst all the other groups prior to backbone fragmentation, we get no structural data on where the oxidation has occurred. And with the shear number of permutations of positions that could be oxidized, we would like to have a general idea where the oxidation is before we go into a wild goose chase.

Have you any experience with ramping up the collision energy in order to favor backbone fragmentation instead of neutral loss through rearrangement?
I'm not sure if i'm following but if you want to prevent your parent ion from fragmention before the collision cell, the collision energy doesn't matter.

Or are you working in SIM mode instead of SRM/MRM? Can you give some information about your instrument?
Thinking more about the problem, losses of small neutrals are inherently more favored energetically than simple backbone cleavage.
NL is a bond breaking (energy in) bond forming (energy out) event whereas backbone cleavage is bond breaking only.
I have never had to do the ramp up expt that you describe.

Does your molecule have 2x OH before CYP treatment?
How many after?

Regards,
JMB
JMB wrote:
Thinking more about the problem, losses of small neutrals are inherently more favored energetically than simple backbone cleavage.
NL is a bond breaking (energy in) bond forming (energy out) event whereas backbone cleavage is bond breaking only.
I have never had to do the ramp up expt that you describe.

Does your molecule have 2x OH before CYP treatment?
How many after?

Regards,
JMB


Hi JMB. I too agree that the neutral loss events are more energetically favorable, that's why this problem we are trying to work around is such a difficult one. After CYP treatment, the molecule is +34 from the original, which accounts for +18 and +16. In total, it would have 4 -OH groups and 1 -NH2 group after metabolism. We are pretty sure the water (+18) added across the single exocyclic alkene that we have, but we are really puzzled about where the +16 occurred. By circumventing neutral loss, we might be able to isolate the +16 in a fragment, so we can determine which "half" the oxidation had occurred. So instead of millions of combinations, we might be able to narrow it down significantly before we try and synthesize it to prove its identity.
Does the pre-CYP cmpd give any [M+H] ?
If so,
1) record MS/MS of [M+H]; go to a nanospray source to accumulate many scans if necessary.
2) record MS/MS of [M+H - H2O]
3) Interpret the spectra as fully as possible.

4) With the post-CYP cmpd, repeat 1) and 2) to give 4a & 4b.

Compare spectra from 1) & 2) with those from 4a & 4b

On a good day, and depending on the structure, you may be able to narrow the oxidation sites to a very few C atoms.
What MW range is the starting cmpd?
JMB wrote:
Does the pre-CYP cmpd give any [M+H] ?
If so,
1) record MS/MS of [M+H]; go to a nanospray source to accumulate many scans if necessary.
2) record MS/MS of [M+H - H2O]
3) Interpret the spectra as fully as possible.

4) With the post-CYP cmpd, repeat 1) and 2) to give 4a & 4b.

Compare spectra from 1) & 2) with those from 4a & 4b

On a good day, and depending on the structure, you may be able to narrow the oxidation sites to a very few C atoms.
What MW range is the starting cmpd?


I see that we can do a step-wise neutral loss CID experiment to determine the fragments in question. Good suggestion. Usually we just get a spread of neutral losses, but I think I can fine tune it a little bit to be more selective.

Our compound is in the low 300s starting.
Some more musings.........
Your molecule is presumably of the form, ignoring the exocyclic DB, CH3(CH2)n(OH)2NH2 where n ~17,18, 19 etc.
I have written -NH2 as a terminal group just for convention.
You may be able to use the NH2 as a marker, since we would generally expect loss of NH3 as well as H2O.
Depending on the resolution settings of your instrument, you may be able to see loss of 17 amu AND 18 amu from the molecular ion.
Assigning the fragments derived solely from the
[MH-NH3]+ ion may help; comparing these with those from the metabolized [MH-NH3]+ may further help to define location of modifications.

Regards
Just some thoughts from the overcautious.... Be careful, there are so many things that can make life difficult!
Firstly, intramolecular rearrangements are very favoured in CID in LC-MS, but this also depends on your instrument (they often require the ion to be in a particular conformation so the reactive groups are approaching one another in the correct way; if you're using an ion-trap with a 30msec collision time, there is all the time in the world for the ion to find itself in the correct configuration).
The rearrangements may be quite extreme (and also highly unlikely-looking in solution chemistry; gas phase chemistry in a vacuum can do odd things).
Secondly, unless you have accurate mass, be cautious about mass differences. For example, if you're already looking at a sodium adduct, you can also gain mass 16 by looking at a potassium adduct instead.
Thirdly, remember that it is surprisingly difficult to tell the difference between an ion with a very labile hydroxyl group, easily lost as water, and an ion that can easily form a cluster-ion with water in the spray chamber. Making conditions in the spray chamber cooler and lower-energy will favour the +18 in both cases. There is a similar problem with ammonium adducts and compounds with very labile amine groups.
Basically when you think you've identified the product, comparing retention time and fragmentation to an authentic standard is essential if you want to be sure it's the right identification.
If you have source fragmentation, it is probably the hydrogen adduct that is fragmenting? If you can find the sodium adduct instead, it probably isn't fragmenting in the source (they're tougher); this can not only check that you've got the correct precursor mass (because you will have fragments that are hydrogen adducts, but the sodium peak only for the parent ion), but also give you something else for which to collect fragments (it is reasonably likely that the sodium and hydrogen adducts will undergo different fragmentations, though in some instruments sodium just gives terrible results!).
Some more musings.........
Your molecule is presumably of the form, ignoring the exocyclic DB, CH3(CH2)n(OH)2NH2 where n ~17,18, 19 etc.
I have written -NH2 as a terminal group just for convention.
You may be able to use the NH2 as a marker, since we would generally expect loss of NH3 as well as H2O.
Depending on the resolution settings of your instrument, you may be able to see loss of 17 amu AND 18 amu from the molecular ion.
Assigning the fragments derived solely from the
[MH-NH3]+ ion may help; comparing these with those from the metabolized [MH-NH3]+ may further help to define location of modifications.

Regards


The issue we have is the order of events I think. If we lose >1 H2O before the backbone fragments, then we basically have no idea which -OH group it was from. If we can control it to 1 H2O, then look at the fragments, we can identify which quadrant it is from and potentially use that to reference the metabolite.

Just some thoughts from the overcautious.... Be careful, there are so many things that can make life difficult!
Firstly, intramolecular rearrangements are very favoured in CID in LC-MS, but this also depends on your instrument (they often require the ion to be in a particular conformation so the reactive groups are approaching one another in the correct way; if you're using an ion-trap with a 30msec collision time, there is all the time in the world for the ion to find itself in the correct configuration).
The rearrangements may be quite extreme (and also highly unlikely-looking in solution chemistry; gas phase chemistry in a vacuum can do odd things).
Secondly, unless you have accurate mass, be cautious about mass differences. For example, if you're already looking at a sodium adduct, you can also gain mass 16 by looking at a potassium adduct instead.
Thirdly, remember that it is surprisingly difficult to tell the difference between an ion with a very labile hydroxyl group, easily lost as water, and an ion that can easily form a cluster-ion with water in the spray chamber. Making conditions in the spray chamber cooler and lower-energy will favour the +18 in both cases. There is a similar problem with ammonium adducts and compounds with very labile amine groups.
Basically when you think you've identified the product, comparing retention time and fragmentation to an authentic standard is essential if you want to be sure it's the right identification.
If you have source fragmentation, it is probably the hydrogen adduct that is fragmenting? If you can find the sodium adduct instead, it probably isn't fragmenting in the source (they're tougher); this can not only check that you've got the correct precursor mass (because you will have fragments that are hydrogen adducts, but the sodium peak only for the parent ion), but also give you something else for which to collect fragments (it is reasonably likely that the sodium and hydrogen adducts will undergo different fragmentations, though in some instruments sodium just gives terrible results!).


It looks like it is the M+H fragmenting. There is very minor fragmentation from the source since M+H is intact. However, if we ramp up the collision energy it just starts to lose the -OH groups as water. I totally understand how difficult it is to favor a less energetically favorable backbone breakage vs. the rearrangement. We generally do not see the sodium adduct in the mass spec experiments that we did. And we do plan to spike to check the retention time once we have a general idea which compound it is to tackle.
sorry, I misunderstood that the loss of water is happening during targeted fragmentation rather than in the spray chamber. I missed what instrument you're using? Are you using a genuine trap, or a triple-like instrument? If it's complex with lots of hydroxyl groups and you don't know which one is (or which ones are) creating your -18 fragment, life is potentially going to be hard. Is there any possibility of scaling up enough for other techniques such as NMR?
lmh wrote:
sorry, I misunderstood that the loss of water is happening during targeted fragmentation rather than in the spray chamber. I missed what instrument you're using? Are you using a genuine trap, or a triple-like instrument? If it's complex with lots of hydroxyl groups and you don't know which one is (or which ones are) creating your -18 fragment, life is potentially going to be hard. Is there any possibility of scaling up enough for other techniques such as NMR?


Oh no worries. The Mass spec we are using is the LTQ Orbitrap XL with EI ionization. It is a triple quad. Our group might be doing some labelling experiments in order to determine where the parent neutral losses are from. Then perhaps we can figure out where the H2Os are coming from for the metabolite. Also, we are trying to scale up the experiment so we will get enough for an NMR. However, once we are at that stage, we might as well go for other techniques such as XRD since the compound should be pretty crystalline. But for now, we will continue to tinker with the MS to see if we can find the condition that would give us more information then it is providing now.
Oh, you're lucky having an LTQ-orbitrap, which is an incredibly flexible ion-trap instrument, rather than a conventional triple. You've probably got multiple modes of CID available, including conventional CID in the low-res trap and HCD in the C-trap, and you might get different results by trying different things. If your molecule tends to lose water, it will probably do that and nothing else in MS2 in normal CID (because obviously the moment it's lost a water, the fragment ion is no longer activated, the LTQ using resonant excitation, so the fragments don't fragment any further, and changing energies won't do a thing). You can use broad-band activation to get more fragments, or go for MS3 etc., but I'm sure you've done that already. Good luck!
14 posts Page 1 of 1

Who is online

In total there are 5 users online :: 0 registered, 0 hidden and 5 guests (based on users active over the past 5 minutes)
Most users ever online was 599 on Tue Sep 18, 2018 9:27 am

Users browsing this forum: No registered users and 5 guests

Latest Blog Posts from Separation Science

Separation Science offers free learning from the experts covering methods, applications, webinars, eSeminars, videos, tutorials for users of liquid chromatography, gas chromatography, mass spectrometry, sample preparation and related analytical techniques.
Subscribe to our eNewsletter with daily, weekly or monthly updates: Food, Environmental, (Bio)Pharmaceutical, Bioclinical, Liquid Chromatography, Gas Chromatography and Mass Spectrometry.

Liquid Chromatography

Gas Chromatography

Mass Spectrometry