Quantitative bioanalysis of a peptide drug - problems

[ian51]

I have applied an LC-MS method for the quantitation of a series of peptide drugs in rat plasma from tk studies. The method employed SPE sample extraction using a hydrophobic interaction polymeric phase, with the sample loaded using 5% TFA (as TFA facilitaes retention of the drug on an ODS HPLC column). LC procedure is a methanol gradient (20% to 90% over 3 mins) with 0.4% TFA as the aqueous mobile phase. Standard ODS column 50 x 3 mm, 5um. Detection is by ESI and SIM mode as compound does not appear to fragment into any useful product ion. Though sensitivity is limited the method worked reasonably well in rat plasma, however, it is a different story in mouse. I am seeing next to nothing in mouse plasma extracts. Spiking extracted mouse plasma gives me peaks and so i've ruled out major signal suppression by mouse plasma as a cause. I believe its a recovery issue, though in one experiment, the first of a number of replicate mouse plasma extrcats yielded peaks which then reduced in intensity dramatically with successive injections. I've also tried TFA protein precipitation prior to SPE to no avail. I would appreciate any advice on why I'm seeing a species difference in application of this method (rapid metabolism of the drug in mouse plasma ? binding of drug to the plasma proteins/pellet ?

Thanks
Ian

[kerri]

Did you try spiking your analytes into the mouse plasma and testing recovery?

If you suspect that the pk is dramatically shorter, you can try shorter timepoints.... if you suspect the extraction recovery is poor, try modifying the spe.

[ian51]

Hi Kerri

Recovery in rat plasma was 70-80%. With mouse plasma, applying the same method, recovery is next to nothing. At this stage I am only working with spiked mouse plasma. I did recover some drug from EDTA collected mouse plasma which which did make me suspect heparin should be avoided as an anticoagulant, but I did recover the drug from heparinised rat plasma. I have analysed the breakthrough from a spiked mouse plasma sample, and the washes used in the SPE procedure but did not detect any drug. I also protein precipitated some spiked mouse plasma samples with 10% (and greater) TFA before taking the supernatants through the SPE procedure, but again low recovery or no peaks at all. As for tk timepoints, I haven't got to that stage yet whilst I still have to resolve the analytical method recovery problem.

Ian

[kerri]

When you test the recovery, how long do you wait to perform the extraction after spiking the peptide in the mouse plasma?

[ian51]

Hi
I spike the mouse plasma and extract immediately. I understand there may be some protease activity, but would be surprised if all the drug was lost so rapidly.

Ian

[kerri]

It's possible that the peptide is protein-bound. You can add DMSO to the plasma before extraction to see if it helps. I know you said that you tried excessive TFA and that didn't help, but DMSO is worth a shot. Possible a DMSO TFA combination would help??

[lhncqt]

Hi,

I have a question: why need to add TFA in precipitation solution?

Thanks,

LHNCQT

[ian51]

As there was a solubility issue when using acetonitrile to precipitate the proteins, I went for an aqueous precipitation agent e.g. zinc sulphate and TCA or TFA. When carrying out SPE alone or following protein precipitation, TFA was needed (as an ion pairing reagent) to facilitate hydrophobic interactions of the peptide drug with the SPE phase. Therefore it made sense to carry out precipitation with TFA and add the supernatant straight to the conditioned SPE plate.

Ultimately I think the problem is heparin. Lastly, I spiked 3 species of plasma with the drug (both EDTA plasma samples and heparinised plasma samples) and managed to recover the drug from all of the EDTA plasma samples but very little from the heparinised plasma samples. As heparin has lots of negatively charged functional groups, and as the peptide has lots of arginine residues with positively charged side groups, I thik the heparin is binding irreversibly to the drug molecules.