LC-MS/MS Method Optimization
We divided target analytes into two groups for liquid chromatography,
referred to as the ‘gonadal steroids’ (17α-hydroxyprogesterone
(17OHP4), androstenedione (AE), testosterone (T), and
progesterone(P4)) and ‘adrenal steroids’ (aldosterone
(ALD), cortisol (F), cortisone (E), corticosterone (B), 11-deoxycortisol
(S), 11-decoxycorticosterone (11DOC), DHEA, and DHEAS) (Table 1). We
sourced standards and internal standards from multiple suppliers
(Supplementary Table 1). We assessed the four gonadal steroids using the
LC-MS/MS method developed previously (6) and detailed in the methods
section. Previous studies have described the separation and quantitation
of five corticosteroids of interest (F, E, B, S, and 11DOC) in blubber
(9, 17). We adapted these methods to incorporate ALD, DHEA, and DHEAS,
for the concurrent measurement of all eight adrenal steroids of
interest.
We conducted optimization using direct injection of individual analytes
(Supp. Table 1) onto an AB Sciex (Framingham, MA) API 4000 QTRAP hybrid
triple quadrupole/linear ion trap mass spectrometer. The Boggs et al.
method (6) used positive mode electrospray ionization (ESI) to identify
mass transitions for corticosteroids (F, E, B, S, and 11DOC). Therefore,
we used matched ionization methods to optimize ALD, DHEA, and DHEAS. We
evaluated fragmentation patterns at varying collision energies until we
identified at least two candidate product ions for each analyte. We used
tuning mode to optimize source (curtain gas, temperature, ion source
gas, interface heater, collision gas, ion spray voltage) and compound
parameters in positive and negative ionization mode (Table 1).
Using a mixture of the eight adrenal analytes, we created a
chromatography method on a Zorbax Eclipse Plus C18 column (150 mm x 2.1
mm, 5 µm particle size) from Agilent with the 1200 Series HPLC system
with a binary pump and autosampler from Agilent (Santa Clara, CA). We
determined compound retention times through multiple-reaction monitoring
(MRM), by monitoring two mass transitions per compound. We selected the
peak with the largest area as the quantitative ion and the second
largest peak as the qualitative ion. The resulting scheduled multiple
reaction monitoring (sMRM) method used a 240-second detection window for
each mass transition, including six internal matched standards (Table 1,
Figure 1).