Separation and quantitative determination of 6α-hydroxycortisol and 6β-hydroxycortisol in human urine by high-performance liquid chromatography with ultraviolet absorption detection

Sensitive and simultaneous quantitation of 6β-hydroxycortisol and cortisol in human plasma by LC-MS/MS coupled with stable isotope dilution method

CYP3A phenotyping provides a means for personalized drug therapy. We focused our attention on the plasma 6β-hydroxycortisol (6β-OHF) to cortisol ratio as an index for CYP3A phenotyping. In the present study, we developed a sensitive and reliable method for the simultaneous determination of 6β-OHF and cortisol in human plasma using high-performance liquid chromatography/tandem mass spectrometry together with picolinylester derivatization or nonderivatization methods and 6β-[9,11,12,12-² H₄ ]hydroxycortisol and [1,2,4,19-¹³ C₄ ]cortisol as internal standards for in vivo CYP3A phenotyping in humans. The lower limits of quantification were 38.513 pg/mL for 6β-OHF and 38.100 pg/mL for cortisol. The relative error and relative standard deviation of the lower limits of quantification were <5% for both methods. The intra-day and inter-day assay reproducibilities of the determined 6β-OHF and cortisol concentrations were consistent with the actual amounts added as relative errors and relative standard deviations for both methods, which were <5.4% and <3.9%, respectively. Both methods were applied for the quantification of plasma 6β-OHF and cortisol concentrations in healthy subjects taking oral contraceptives. The absolute concentrations and time course of 6β-OHF and cortisol were found to be consistent when measured using the 2 methods. The ratio as an index for in vivo CYP3A activity decreased after 21 days of taking oral contraceptives for both methods. To the best of our knowledge, this is the first study reporting the detailed investigation of accuracy and precision in the simultaneous measurement of 6β-OHF and cortisol in human plasma using liquid chromatography/tandem mass spectrometry coupled with stable isotope dilution method, which can be applied to CYP3A phenotyping.

Detection and characterization of prednisolone metabolites in human urine by LC-MS/MS

Glucocorticosteroids are prohibited in sports when used by systemic administrations (e.g. oral), whereas they are allowed using other administration ways. Strategies to discriminate between administrations routes have to be developed by doping control laboratories. For this reason, the metabolism of prednisolone (PRED) was studied using liquid chromatography coupled to tandem mass spectrometry. A single oral (10 mg) dose of PRED was administered to two healthy male volunteers. Urine samples were collected up to 6 days after administration. Samples were hydrolyzed with β-glucuronidase and subjected to liquid-liquid extraction with ethyl acetate in alkaline conditions. The extracts were analyzed by liquid chromatography coupled to tandem mass spectrometry. Precursor ion scan methods (m/z 77, 91, 105, 121, 147 and 171) in positive ionization and neutral loss scan methods (76 and 94 Da) in negative ionization modes were applied for the open detection of PRED metabolites. Using these methods, PRED parent compound plus 20 metabolites were detected. PRED and 11 metabolites were characterized by comparison with standards of the compounds (PRED, prednisone, 20β-dihydro-PRED and 20α-dihydro-PRED, 20β-dihydro-prednisone and 20α-dihydro-prednisone, 6β-hydroxy-PRED and 6α-hydroxy-PRED, 20β isomers and 20α isomers of 6β,11β,17α,20,21-pentahydroxypregnan-1,4-diene-3-one, 6α,11β,17α,20β,21-pentahydroxypregnan-1,4-diene-3-one and Δ(6) -PRED). Using mass spectrometric data, feasible structures were proposed for seven of the remaining nine detected metabolites, including several 6-hydroxy-metabolites. Eleven of the characterized metabolites have not been previously described. Maximum excretion rates for PRED metabolites were achieved in first 24 h; however, most of the metabolites were still detectable in the last collected samples (day 6).

Use of endogenous cortisol 6β-hydroxylation clearance for phenotyping in vivo CYP3A activity in women after sequential administration of an oral contraceptive (OC) containing ethinylestradiol and levonorgestrel as weak CYP3A inhibitors

The present study was undertaken to evaluate the time courses of in vivo cytochrome P450 3A (CYP3A) inhibition in four healthy women after sequential administration of an oral contraceptive (OC) containing ethinylestradiol and levonorgestrel, using 6β-hydroxylation clearance of endogenous cortisol (CLm(6β)) as a new index for CYP3A phenotyping. The 6β-hydroxylation clearance (CLm(6β)) was followed every 2h from 9:00 or 11:00 to 17:00 on days 0 (baseline), 1, 2, 21, and 28 during a single menstrual cycle. The serum concentrations of endogenous estradiol and progesterone were also measured. The time course data of CLm(6β) clearly demonstrated 43-64% inhibition of CYP3A activity in women taking a low daily dose of the OC for 21days. The average CLm(6β) levels that were suppressed by the OC in four women were extremely low (0.60-1.23mL/min) compared with the normal CLm(6β) range (1.5-3.5mL/min) that was obtained from 49 healthy subjects in our previous study. The in vivo inhibitory potencies (43-64%) obtained in this study were stronger than expected from reported in vitro studies (∼20%). Furthermore, it would take at least seven days to return to the baseline activity of CYP3A after discontinuation of the OC. The results presented here should provide important information on the inhibitory effect of OC on the CYP3A activities in women, which are involved in the metabolism of a number of drugs with a narrow therapeutic range.

Detection and characterization of triamcinolone acetonide metabolites in human urine by liquid chromatography/tandem mass spectrometry after intramuscular administration

RATIONALE

Glucocorticosteroids are prohibited in sports when used by systemic administrations (e.g. intramuscular, IM), whereas they are allowed using other ways of administration. Strategies to discriminate between administrations routes have to be developed by doping control laboratories. For this reason, the metabolism of triamcinolone acetonide (TA), one of the most used glucocorticosteroids, was studied using liquid chromatography coupled to tandem mass spectrometry (LC/MS/MS).

METHODS

Urine samples obtained after IM administration of TA were analyzed using two sample treatments: (a) hydrolysis with β-glucuronidase enzymes and liquid-liquid extraction under alkaline conditions, and (b) liquid-liquid extraction under acidic conditions. The extracts were analyzed by LC/MS/MS.

RESULTS

TA, commercially available metabolites (6β-hydroxytriamcinolone acetonide, 6β-OH-TA, and triamcinolone), and their C20-reduced derivatives showed characteristic fragmentation behavior. Besides common product ions and neutral losses for corticosteroids containing fluorine, additional characteristic neutral losses (58 Da, loss of acetone; 44 Da, loss of acetaldehyde) were observed in positive electrospray ionization. Based on that behavior, two complementary approaches were applied to detect TA metabolites: (a) open detection by precursor ion and neutral loss scan methods and (b) targeted detection by selected reaction monitoring methods (SRM) containing theoretical ion transitions of the potential metabolites. Two main compounds, TA and 6β-OH-TA, and nine minor potential metabolites, were detected by open screening methods. Using SRM, two additional metabolites were detected. Some of the metabolites were characterized using reference standards and, for the rest of metabolites, feasible structures were proposed based on mass spectrometric data.

CONCLUSIONS

Metabolites resulting from hydroxylation in C-6, oxidation of the 11-hydroxyl group, reduction of the Δ(4) double bond and oxidation of the side chain were detected. Some of them have not been previously described. Excretion profiles of the detected metabolites after IM administration are presented.

Determination of total concentration of chemically labeled metabolites as a means of metabolome sample normalization and sample loading optimization in mass spectrometry-based metabolomics

For mass spectrometry (MS)-based metabolomics, it is important to use the same amount of starting materials from each sample to compare the metabolome changes in two or more comparative samples. Unfortunately, for biological samples, the total amount or concentration of metabolites is difficult to determine. In this work, we report a general approach of determining the total concentration of metabolites based on the use of chemical labeling to attach a UV absorbent to the metabolites to be analyzed, followed by rapid step-gradient liquid chromatography (LC) UV detection of the labeled metabolites. It is shown that quantification of the total labeled analytes in a biological sample facilitates the preparation of an appropriate amount of starting materials for MS analysis as well as the optimization of the sample loading amount to a mass spectrometer for achieving optimal detectability. As an example, dansylation chemistry was used to label the amine- and phenol-containing metabolites in human urine samples. LC-UV quantification of the labeled metabolites could be optimally performed at the detection wavelength of 338 nm. A calibration curve established from the analysis of a mixture of 17 labeled amino acid standards was found to have the same slope as that from the analysis of the labeled urinary metabolites, suggesting that the labeled amino acid standard calibration curve could be used to determine the total concentration of the labeled urinary metabolites. A workflow incorporating this LC-UV metabolite quantification strategy was then developed in which all individual urine samples were first labeled with (12)C-dansylation and the concentration of each sample was determined by LC-UV. The volumes of urine samples taken for producing the pooled urine standard were adjusted to ensure an equal amount of labeled urine metabolites from each sample was used for the pooling. The pooled urine standard was then labeled with (13)C-dansylation. Equal amounts of the (12)C-labeled individual sample and the (13)C-labeled pooled urine standard were mixed for LC-MS analysis. This way of concentration normalization among different samples with varying concentrations of total metabolites was found to be critical for generating reliable metabolome profiles for comparison.