Analysis of Vitamin D in blood by LC-MSMS

Hi dear community members.

I need your suggestions about analyzing Vitamin D in blood by LC-MSMS. I tried a method with below parameters:

 

Sample preparation:

Transfer 100 µl of reconstituted calibration standard (prepared by dissolving the lyophilized std. in 3 ml deionized water) into LC vial. Add to it 350 µl of the precipitating solution (Acetonitrile: Methanol 95:5, v/v %). Vortex for 20-30 seconds. Centrifuge for 15-20 min at 4700 rpm. The supernatant is injected to LC-MSMS (Source: App Note TN-1055, Ultra-fast LC/MS/MS Analysis of 25-OH Vitamin D2 and D3 from Serum, Phenomenex).

Mobile Phase Gradient:

Time (min)

A(%)-0.1 % Formic acid in water

B(%)-0.1 % Formic acid in methanol

0

15

85

3.2

15

85

3.21

2

98

3.80

2

98

3.81

15

85

 

Flow rate 0.5 ml/min. Stop time at 5 min.

Column: RRHD Eclipse Plus C18, 2.1x50mm, 1.8u (p/n 959757-902)

The instrument is Agilent 1260 HPLC with 6460 MSMS with AJS ESI source. The MRM transitions are attached.

As you can see from the attached screenshots in the resultant chromatogram of the calibrator, I got peaks for both MRM transitions (401.3 >> 383.2, 401.3 >> 365.3 both at collision energy 20 V) of vitamin D3 i.e a peak appears at the same retention time in both MRM transitions. However, I could not get any normal peak in both transitions of vitamin D2 (413.3 >> 395.3, 413.3 >> 355.2 both at collision energy 4 V).

Assuming that I had a good peak for vitamin D3 I decided to make calibration. I prepared samples for 6 levels of calibrators for Vitamin D3 as per sample preparation method described above. However when I analyzed the data in the Quant, it turned out that the responses were chaotic (see the attached screenshots), they are not increasing as expected.

So I have 2 questions:

  • What could be done to get the peak of vitamin D2?
  • Why responses of Vitamin D3 is not linear with calibration levels. Is this peak really belongs to vitamin D3.
attachments.zip
Parents
  • I'm not familiar with analysing Vitamin D specifically, but looking at the structures of the two compounds I'd be surprised if they had such different collision energies. The application note you reference has a collision energy of 20 for D2 and 31 for D3 - this is on a SCIEX triple quad so it won't be surprising to have different numbers on a 6460, but I would expect D2 and D3 to have similar collision energies to each other. Did you use the MRM optimisation tool to get your transitions?

     

    Another thing you could look at for improving your calibration curve is using an internal standard to account for any extraction variability, but I think you need to sort out the transitions first.

Reply
  • I'm not familiar with analysing Vitamin D specifically, but looking at the structures of the two compounds I'd be surprised if they had such different collision energies. The application note you reference has a collision energy of 20 for D2 and 31 for D3 - this is on a SCIEX triple quad so it won't be surprising to have different numbers on a 6460, but I would expect D2 and D3 to have similar collision energies to each other. Did you use the MRM optimisation tool to get your transitions?

     

    Another thing you could look at for improving your calibration curve is using an internal standard to account for any extraction variability, but I think you need to sort out the transitions first.

Children
No Data
Was this helpful?