Urinary Marker of Oral Pregnenolone Administration Christophe Saudan ∗, Aurelien´ Desmarchelier, Pierre-Edouard Sottas, Patrice Mangin, Martial Saugy

Home , Androstenol, Etiocholanolone, Pregnanediol

Steroids 70 (2005) 179–183

Urinary marker of oral pregnenolone administration Christophe Saudan ∗, Aurelien´ Desmarchelier, Pierre-Edouard Sottas, Patrice Mangin, Martial Saugy

Laboratoire Suisse d’Analyse du Dopage, Institut Universitaire de M´edecine L´egale, Rue du Bugnon 21, 1005 Lausanne, Switzerland Received 13 September 2004; received in revised form 29 November 2004; accepted 3 December 2004

Abstract

Pregnenolone (PREG) can potentially be abused by athletes to maintain an equilibration of the steroidal environment after sex steroids administrations. Five men volunteers orally ingested 50 mg PREG to determine optimal urinary markers for detection of this steroid. Our ﬁndings show that ingestion of PREG has no signiﬁcant effects on the testosterone/epitestosterone (T/E) and testosterone/luteinizing hormone (T/LH) ratios, whereas variable changes on the carbon isotopic values of three T metabolites: androsterone, etiocholanolone, 5␤-androstane- 3␣,17␤-diol (5␤-androstanediol) together with 16(5␣)-androsten-3␣-ol (androstenol) and 5␤-pregnane-3␣,20␣-diol (pregnanediol) have been observed. The difference between the carbon isotopic values (δ13C-values) of androstenol and pregnanediol is potentially the most reliable marker of exogenous PREG administration in males. For all subjects, the differences differ by 3.0‰ or more over a period of about 10 h and for both of them the detection window for positivity is extended over 40 h. © 2005 Elsevier Inc. All rights reserved.

Keywords: Pregnenolone; Urinary marker; Isotope ratio mass spectrometry (IRMS); Doping control

1. Introduction in humans have shown significant anti-inflammatory effects in rheumatoid arthritis and anti-fatigue effects with The biosynthesis of steroid hormones starts with choles- obvious improvement in performance [3]. In the case of terol from which the sex steroids, glucosteroids and min- large excess of sex steroids, it is supposed that PREG eralosteroids all derive [1]. Pregnenolone (3␤-hydroxy-5- would help in maintaining an equilibration of steroidal pregnen-20-one, PREG) is the major precursor for the environment by furnishing substances along other paths steroids hormones and is regulated by pituitary hormones, [3]. such as luteinizing hormone (LH) and follicle-stimulating The purpose of this study is to investigate first the effect of hormone (FSH) in ovaries and testes. PREG is a freely a single oral PREG administration on known indirect markers available compound in the United States and is associated of steroids in urine. As PREG is a precursor of the sex steroids, with other “pro-hormones” such as DHEA or 4-androstene- the administration of this substance may potentially increase 3,17-dione in nutritional supplements. Since its ingestion the concentration of testosterone and therefore may result has not been perceived to contribute to androgen syn- in a substantial increase of the testosterone/epitestosterone thesis, this steroid has not been included so far on the (T/E) and T/LH ratios. Indeed, the T/E and T/LH ratios have List of Prohibited Substances of the World Anti-Doping been found to be reliable urinary markers of exogenous T Agency (WADA) [2]. Nevertheless, studies with PREG administration in males [4–6]. Gas chromatography/combustion/isotopic ratio mass spectrometry (GC/C/IRMS) analysis of three T metabo- ∗ Corresponding author. Tel.: +41 21 314 7106; fax: +41 21 314 7333. lites (sex steroids pathway): androsterone, etiocholanolone, E-mail address: [email protected] (C. Saudan). 5␤-androstane-3␣,17␤-diol together with 16(5␣)-androsten-

3␣-ol (16-ene steroids pathway) and 5␤-pregnane-3␣,20␣- diol (a metabolite of progesterone) has been also applied to search for urinary markers of PREG administration. Al- though PREG is a precursor of all steroids, it is expected that the administration of this compound will not affect similarly the isotopic values of the investigated steroids [7].

2. Materials and methods

2.1. Subjects Fig. 1. Urinary T/E ratio of five subjects (S1–S5) before and after oral ad- Five Caucasian healthy male volunteers (ages 23–50 ministration of 50 mg PREG at t = 0. The horizontal dashed line displays T/E years, mean = 34.4, S.D. = 9.8) living in Switzerland ratio = 4. were self-administered a 50 mg capsule of pregnenolone (Metabolic Response ModifierTM, San Clemente, CA, USA). 3. Results and discussion The study was in accordance with the Helsinki Declaration of 1975 and all of the subjects gave their written informed 3.1. Urine steroid profile consent. Baseline urine samples were obtained before initial administration, and subsequent spot urine samples were col- As shown in Fig. 1, no T/E ratio values approached the lected over a period of 80 h after the capsule administration. 4:1 threshold that suggests testosterone abuse. This threshold Each urine sample was divided into 20-ml flasks and stored was recently adopted with the need of submitting the sam- without additives at −20 ◦C until analysis. ple to IRMS analysis if the urinary T/E ratio value is equal or greater than 4:1 [10]. The basal T/E ratios for all subjects of this study are ranging from 0.78 to 1.85, and do not 2.2. Analytical methods vary significantly upon PREG administration (Pmin = 0.22). In addition, we have not observed any substantial increase of The T/E ratio was determined by GC–MS according to a testosterone in the urine after adjustment for specific grav- published procedure [8]. The T and E calibration curves were ity [11]. The mean urinary concentration of T glucuronide 2 linear (R = 0.999) between 5 and 150 ng/ml. throughout the study was of 61 ± 17, 79 ± 27, 22 ± 6, 41 ± 17 The isotope ratio mass spectrometry (IRMS) method and 34 ± 21 ng/ml for subjects S1, S2, S3, S4 and S5, re- δ13 and experimental conditions for the urinary C-value spectively. Although PREG is a precursor of T, our results ␣ ␣ determination of 16(5 )-androsten-3 -ol (androstenol), indicate that only a small amount of PREG is converted into ␤ ␣ ␣ 5 -pregnane-3 ,20 -diol (pregnanediol), androsterone, T. This agrees with a report on oral administration of de- ␤ ␣ ␤ ␤ etiocholanolone and 5 -androstane-3 ,17 -diol (5 - hydroepiandrosterone (DHEA), which originate from PREG androstanediol) have been described [7]. For the IRMS metabolism by the action of P450c17 in humans [12]. In this measurements of PREG contained in the capsule, the steroid report, it has been shown that about 1.5% DHEA is converted was extracted according to a published method for nutritional into T and that a single oral dose of DHEA do not increase supplements [9], excepted that the residue was derivatised the T/E ratio in a group of nine healthy males. with 50 ␮l of pyridine and 100 ␮l of acetic anhydride for 1 h ◦ Similarly to the T/E ratio, oral administration of 50 mg at 60 C. Finally, the presence of PREG in the extracts was PREG has not lead to significant time changes in urinary verified by GC/MS prior to GC/C/IRMS analysis with the T/LH ratio (data not shown). experimental conditions used for pregnanediol [7]. The urine concentration of LH was determined on a Cobas 3.2. Isotopic urinary markers of PREG administration Core II instrument (Roche Pharma AG, Reinach, Switzer- land) with the LH EIA II kit purchased from Roche (Basel, Isotope ratio mass spectrometry allows measurements of Switzerland). slight differences in the carbon isotope ratio (13C/12C) of the exogenous and endogenous steroids [13–15]. Synthetic 2.3. Statistical method steroids are generally produced from precursors derived from plants with low 13C content, whereas the 13C and 12C con- Global statistical analysis for all subjects as a function tent in the natural endogenous form depends on the isotopic of time was carried out by MATLAB (Statistics Toolbox, carbon composition of the food diet and is influenced by addi- Version 3.0) using the non-parametric one-way analysis of tional effects of human biological processing. As illustrated variance (Kruskal–Wallis test). Significant increase above the in Fig. 2, the δ13C-values of the investigated steroids prior basal levels with P < 0.05 was considered. to oral administration of PREG are ranging from −22.77 to C. Saudan et al. / Steroids 70 (2005) 179–183 181

Fig. 2. Urinary δ13C-values of androsterone (), etiocholanolone (), pregnanediol (᭹), 5␤-androstanediol () and androstenol () of ﬁve subjects (S1–S5) before and after oral administration of 50 mg PREG at t = 0. The dash lines with δ13Cof−30.50‰ represent the value of the isotope ratio of PREG in the capsule.

−24.20‰, whereas the δ13C-value of PREG in the capsule Global statistical analysis for all subjects as a function of time is −30.50‰ (n = 3, S.D. = 0.13‰). Therefore, these findings displays no significant changes for androsterone after oral confirm the lower 13C content of the exogenous steroid with administration of 50 mg PREG, whereas 5␤-androstanediol respect to endogenous compounds and show that a variation and etiocholanolone were found to significantly change from up to 7.7‰ may be expected upon oral administration of the 4 to 31 h and from 6 to 12 h post-administration, respectively. capsule. Similarly to 5␤-androstanediol, the isotopic ratio change of The time course of urinary isotopic values of 16(5␣)- androstenol was significant for more than 25 h after PREG androsten-3␣-ol (androstenol) together with 5␤-pregnane- administration (from 6 to 34 h post-administration). 3␣,20␣-diol (pregnanediol) and three testosterone metabo- Despite the significance of the isotopic ratio changes as lites: androsterone, etiocholanolone, 5␤-androstane-3␣,17␤- a function of time, it should be emphasized that the stan- diol (5␤-androstanediol) for all subjects are shown in Fig. 2. dard deviation on the measurement is of about 0.4‰ [7] and The spot urines were selected for IRMS analyses to have that the differences with the mean basal values are less than ␤ similar excretion time among the subjects (S.D.max = 2.5 h). 2.5, 1.1 and 1.9‰ for 5 -androstanediol, etiocholanolone 182 C. Saudan et al. / Steroids 70 (2005) 179–183 and androstenol, respectively. As previously observed [7,16], subjects (Fig. 2). However, it is conceivable that concomi- the δ13C-values of pregnanediol are strongly affected by tant oral intakes of PREG and sex steroids (testosterone, administration of PREG. Differences with the mean basal androstenedione, DHT) or DHEA would strongly affect the values higher than 4.0‰ were detected at about 10 h post- isotopic ratios of pregnanediol together with all T metabo- administration. Subject 1 displays the largest variation for lites [7,16,18], with limited changes for androstenol. Indeed, the δ13C-values of pregnanediol with a minimum value of testosterone and other compounds, including epitestosterone −29.6‰ (Fig. 2). As the δ13C-value of PREG in the capsule and DHEA, may be excluded as precursors of androstenol is of −30.50‰, the results obtained with subject S1 tend to [19,20]. It has been shown recently that androstenol is syn- demonstrate that pregnanediol contained in the urines from thesized from PREG through the 16-ene steroids pathway 17 to 25 h post-administration originates almost completely by utilizing P450c17 and the enzymes of the steroidogenic from the bio-transformation of the exogenous steroid. For the pathway [19,21]. other subjects, it is likely that both exogenous and endogenous PREG with different ratio have been metabolized into pregnanediol. 4. Conclusion The use of pregnanediol as an endogenous reference compound (ERC) for IRMS analysis is worth discussing. Al- Both T/E and T/LH ratios failed to be sensitive markers though administration of steroids such as DHEA, testos- of oral administration of PREG. A reliable urinary marker terone, androstenedione and DHT will not affect the isotopic of doping with this compound is obtained by applying a ratio of this endogenous reference compound [7,13,17], our GC/C/IRMS method [7]. The difference between the carbon findings demonstrate that administration of PREG leads to isotopic values (δ13C-values) of androstenol and pregnane- significant changes of the δ13C-values of pregnanediol. Thus, diol is found to clearly indicate exogenous PREG administra- the use of pregnanediol and most likely of 5␤-pregnane- tion and is likely to specifically detect doping with PREG in 3␣,17␣,20␣-triol (metabolite of 17-hydroxyprogesterone) as the case of concomitant intakes of PREG and a sex steroid or ERC will be seriously compromised if PREG has been orally DHEA. The positivity window of this urinary marker shows ingested. In such instance, the use of androstenol as ERC may interindividual variability, but is at least 10 h for the dose be suggested since smaller isotopic ratio variations compared ingested. to pregnanediol are observed (Fig. 2). Therefore, the large variation of pregnanediol isotopic ratios after PREG administration has been potentialized by determining the difference Acknowledgements between the δ13C-values of androstenol and pregnanediol as a urinary marker of doping with PREG. Fig. 3 shows that all We thank Carine Schweizer and Norbert Baume for sci- subjects would be reported as consistent with the administra- entific advice. Dr. Neil Robinson and Laurent Perrenoud are 13 12 tion of PREG over a period of about 10 h, since the C/ C acknowledged for their technical assistance. This study was value measured for the metabolite differs significantly, i.e. supported by a grant from the Swiss Federal Office of Sport by 3.0‰ or more from that of the urinary reference steroid (FOSPO). chosen [10]. For subjects S1 and S3, the detection window for positivity extends over 40 h. Using androsterone as ERC would provide best evidence of PREG administration, since References androstenol is found to vary by more than 1.5‰ in some [1] Miller WL. Molecular biology of steroid hormone synthesis. Endocr Rev 1988;9:295–318. [2] WADA, List of Prohibited Substances, 2004, http://www.wada- ama.org/. [3] Roberts E. Pregnenolone — from Selye to Alzheimer and a model of the pregnenolone sulfate binding site on the GABAA receptor. Biochem Pharmacol 1995;49:1–16, and references therein. [4] Palonek E, Gottlieb C, Garle M, Bjorkhem¨ I, Carlstrom¨ K. Serum and urinary markers of exogenous testosterone administration. J Steroid Biochem Mol Biol 1995;55:121–7. [5] Kicman AT, Brooks RV, Collyer SC, Cowan DA, Nanjee MN, Southan GJ, et al. Criteria to indicate testosterone administration. Br J Sp Med 1990;24:253–64. [6] Peng SH, Segura J, Farre´ M, Gonzalez´ JC, de la Torre X. Plasma and urinary markers of oral testosterone undecanoate misuse. Steroids 2002;67:39–50. Fig. 3. Difference between androstenol and pregnanediol δ13C-values [7] Saudan C, Baume N, Mangin P, Saugy M. Urinary analysis of (δ13C-values) of five subjects (S1–S5) after oral administration of 50 mg 16(5␣)-androsten-3␣-ol by gas chromatography/combustion/isotope PREG at t = 0. The dashed box includes the δ13C-values ranging from 3.0 ratio mass spectrometry: implications in anti-doping analysis. J Chro- to 5.0‰. matogr B 2004;810:157–64. C. Saudan et al. / Steroids 70 (2005) 179–183 183

[8] Nolteernsting E, Geyer H, Mareck-Engelke U, Schanzer¨ W, Donike [15] Aguilera R, Chapman TE, Starcevic B, Hatton CK, Catlin DH. Per- M. Standardization of the T/E determination by deuterated internal formance characteristics of a carbon isotope ratio method for de- standards. In: Donike M, Geyer H, Gotzmann A, Mareck-Engelke tecting doping with testosterone based on urine diols: controls and U, editors. Recent advances in doping analysis (2), Proceedings athletes with elevated testosterone/epitestosterone ratios. Clin Chem of the Manfred Donike Workshop on Dope Analysis, 1994. p. 2001;47:292–300. 113–20. [16] Ferry M, Mathurin JC, Becchi M, de Ceaurriz J. Influence of preg- [9] Delbeke FT, van Eenoo P, van Thuyne W, Desmet N. Prohormones nenolone administration on IRMS analysis. In: Schanzer¨ W, Geyer and sport. J Steroid Biochem Mol Biol 2003;83:245–51. H, Gotzmann A, Mareck-Engelke U, editors. Recent advances in [10] WADA Technical Document: Reporting and evaluation guidance doping analysis (8), Proceedings of the Manfred Donike Workshop for testosterone, epitestosterone, T/E ratio and other endogenous on Dope Analysis, 2000. p. 209–13. steroids, 2004, http://www.wada-ama.org/. [17] Ayotte C, Gouldreault D, Lajeunesse A, Cleroux´ M, Richard Y, [11] Baume N, Avois L, Schweizer C, Cardis C, Dvorak J, Cauderay M, Charlebois A, et al. GC/C/IRMS and GC/MS in “natural” steroids et al. [13C]Nandrolone excretion in trained athletes: interindividual testing. In: Schanzer¨ W, Geyer H, Gotzmann A, Mareck-Engelke U, variability in metabolism. Clin Chem 2004;50:355–64. editors. Recent advances in doping analysis (9), Proceedings of the [12] Dehennin L, Ferry M, Lafarge P, Per´ es` G, Lafarge JP. Oral ad- Manfred Donike Workshop on Dope Analysis, 2001. p. 133–43. ministration of dehydroepiandrosterone to healthy men: alteration of [18] Maˆıtre A, Saudan C, Mangin P, Saugy M. Urinary analysis of the urinary androgen profile and consequences for the detection of four testosterone metabolites and pregnanediol by gas chromatog- abuse in sport by gas chromatography–mass spectrometry. Steroids raphy/combustion/isotope ratio mass spectrometry after oral admin- 1998;63:80–7. istrations of testosterone. J Anal Toxicol 2004;28:426–31. [13] Shackleton CHL, Phillips A, Chang T, Li Y. Confirming testosterone [19] Soucy P, Lacoste L, Luu-The V. Assessment of porcine and human administration by isotope ratio mass spectrometric analysis of urinary 16-ene-synthase, a third activity of P450c17, in the formation of an androstanediols. Steroids 1997;62:379–87. androstenol precursor. Eur J Biochem 2003;270:1349–55. [14] Shackleton CHL, Roitman E, Phillips A, Chang T. Androstane- [20] Smals AGH, Weusten JJAM. 16-ene-steroids in the human testis. J diol and 5-androstenediol profiling for detecting exogenously admin- Steroid Biochem Mol Biol 1991;40:587–92. istered dihydrotestosterone, epitestosterone, and dehydroepiandros- [21] Dufort I, Soucy P, Lacoste L, Luu-The V. Comparative biosynthetic terone: potential use in gas chromatography isotope ratio mass spec- pathway of androstenol and androgens. J Steroid Biochem Mol Biol trometry. Steroids 1997;62:665–73. 2001;77:223–7.