Results and discussion
Free and total steroids were quantified for male and female urine. Steroids were comprehensively analyzed using the Li and H methods. As discussed in the previous paper, 3-OH steroids severely undergo dehydration when ionized in positive ion mode14. Furthermore, additional OH group(s) in the backbone, such as7-OH DHEA and 7-OH P5, further promote dehydration during ionization and those are difficult to detect by conventional LC/ESI-MS/MS. The adduction of Li ions to steroid molecules in the ionization could protect them from dehydration, resulting in an increase in ion intensity of the precursor ion as a lithiated form and efficient generation of specific fragment ions, especially, for 3-OH steroids14 (Figure 2 and Figure S2). For 3-ketosteroids, the ”H method” could be used because they do not contain such hydroxyl groups. Furthermore, the analytical settings of the H-method and the Li-method are very similar, so they can be operated in a consecutive manner with a regular LC-MS/MS, allowing for a more comprehensive analysis of multi-class steroids in bio-samples. Figure 2 compares MRM chromatograms of free and total 7-OH DHEA and 7-OH P5 from female and male specimens; 7-OH P5 was only observed in total steroids of female specimen, while 7-OH DHEA was observed in free and total steroids prepared from female and male samples. These two steroids were first identified by the Li method in human urine. When these steroids were measured by the H-method, 7-OH P5 was barely observed and 7-OH DHEA was likely to be observed as a shouldered peak, but the retention time and peak height ratio of the two transitions differed significantly from the standard one (Figure 2, right panel). This could be partly attributed to instability of the protonated ion, MH+ or (M-H2O+H)+ ions, in the ion source. The adduction of a Li ion to steroid in the gas phase should protect it from the dehydration or degradation during ionization. Indeed, the sensitization ratios of 7-OH-DHEA and 7-OH-P5 were 7.1 and 11.1, respectively (Figure 2).
19 free and 29 total steroids were identified in the urine of men and women. The amounts of each steroid was normalized by creatinine values obtained separately for male and female samples (Table S5). The relative composition of free steroids was different from the total steroids, the majority of which consists of the conjugated forms. The predominant urinary steroids (conjugated forms) were TH-COL, TH-COR, APH, COB, THB, An, DHEA, and 11-OH-An. Female had relatively higher contents of 7-OH-P5, 16OH-E1, E3, P5, and 7-OH-DHEA than male, while male had higher contents of DOC, AT, TH-DOC, TE, COR, COL THB, 11-OH-An and An, a result that clearly reflects gender differences (Figure 3). Since 7-OH-P5, 16OH-E1, E3, P5, 7-OH-DHEA, AT, TE, 11-OH-An and An are sex steroids, it is normal for significant differences between male and female urine samples. The stress-related corticosteroids DOC, TH-DOC, COR, COL and THB were common in male17, however, this result needs to be confirmed with more specimens. Note that the amounts of each steroid were obtained for the mixture of three specimens of urine and their levels were almost in the ranges of those reported previously2,18-25 (Figure S5), which were obtained for individuals’ urine or 24-hours pooled urine22, while E2, THB were observed with higher levels than those reported previously. It could be ascribable to inter-day or individual variation of steroid metabolism.
The high sensitivity of the Li method also revealed the profile of free steroids in urine (about 1.2% of total steroid levels), although their physiological function is unclear. It is worth mentioning that since the method allowed the reliable quantification of 7-OH DHEA, which has been reported to be involved in “doping”, it will be useful for its monitoring26.