Evaluation of human hepatocyte incubation as a new tool for metabolism study of androstenedione and norandrostenedione in a doping control perspective

doi:10.1016/S1570-0232(02)00444-0

Journal of Chromatography B

Volume 780, Issue 1, 15 November 2002, Pages 145-153

https://doi.org/10.1016/S1570-0232(02)00444-0 Get rights and content

Abstract

Human hepatocyte incubations were used to study the metabolism of precursors of testosterone and nortestosterone and to evaluate qualitatively the correlation between in vitro and published in vivo urinary metabolic profiles. Both phase I and phase II biotransformations were observed in vitro: oxidoreduction at C-3 and C-17, reduction at C-4,5, hydroxylation at C-6β and C-16, glucuronidation and sulfation. All major metabolites detected in urine following oral administration of androstenedione and norandrostenedione were present in human hepatocyte incubations. The good correlation between in vitro and in vivo metabolic profiles indicates that hepatocyte incubations can be a useful tool to identify and characterize metabolites that could be potential urinary markers for detection of steroid abuse by athletes.

Introduction

The Medical Commission of the International Olympic Committee (IOC) prohibits the use by athletes of over 100 doping agents classified in five categories including the anabolic agents [1]. From synthetic anabolic steroids (e.g. stanozolol) to natural hormones (e.g. testosterone), athletes use different types of Anabolic–Androgenic Steroids (AAS), mainly injectables, to increase their muscular mass and to enhance the intensity and duration of training sessions [2].

The misuse of anabolic steroids is detected by evaluating the presence of banned substances or their metabolites in urine samples. Metabolism studies of anabolic steroids are usually performed in vivo in humans. After oral or intra-muscular administration of the doping agent, urine samples are collected over a certain period of time. Standard extraction and derivatization methods are followed by GC–MS analysis for detection and quantification [3], [4], [5]. Good correlation between in vivo studies and so-called urinary positive steroid profiles is observed. However, urine samples contain many interfering substances complicating the detection and identification of new urinary markers. Moreover, metabolites of interest are often present at low concentrations in the urine, leading to detection problems and characterization difficulties.

For the past 20 years, pharmaceutical and academic research groups have been using in vitro systems to study the metabolism of drugs and potential drug candidates to ultimately predict their in vivo profile in humans [6], [7], [8]. In vitro incubations are usually performed in sub-cellular systems such as microsomes under oxidative, reductive or phase II conjugation conditions or in whole cells such as hepatocytes. These in vitro systems provide the opportunity to form higher concentrations of metabolites, detect intermediates, isolate metabolites for further characterization, evaluate kinetics, study tissue specificity and maximize safety by reducing the need for in vivo studies. Veterinary agencies use in vitro bovine hepatocyte incubations to study metabolism of steroids, considered as growth promoter agents in animal production [9], [10], [11]. To our knowledge, limited work using such in vitro tools has been done over the years in the human doping control field. Rendic et al. studied the implication of different cytochrome P450s in the oxidative metabolism of steroids using human recombinant enzymes [12]. In vitro incubations could be useful tools for the identification and characterization of new metabolites as potential urinary markers for the detection of steroid abuse.

Androstenedione and norandrostenedione (Fig. 1) were selected to evaluate the potential of in vitro incubations to study the metabolism of anabolic androgenic steroids in humans. These steroids are precursors of testosterone and nortestosterone respectively and are prohibited by the medical commission of the IOC [1]. Over the past 5 years, scientific studies have been done to determine the in vivo metabolism profiles of these steroids [13], [14], [15], [16]. Following oral administration, similar biotransformations were observed for both compounds: reduction of double bond at C-4,5, reduction of ketones at C-3 and C-17, hydroxylation and conjugation of phase I metabolites with glucuronic acid or sulfate groups. Knowing that these steroids are likely to be administered per os as dietary supplements, their absorption by the digestive track system should lead to first pass metabolism by the liver [17]. Therefore, hepatocytes isolated from human liver represent a good alternative to in vivo studies to look at the metabolism profile of such compounds. In order to confirm the usefulness of human hepatocyte incubations, we first studied the in vitro metabolic profiles of androstenedione and norandrostenedione and evaluated qualitatively the correlation with published in vivo urinary metabolic profiles. Then, the major biotransformations observed in vitro were compared to known in vivo metabolic pathways.

Section snippets

Standards and chemicals

All the following steroid standards were purchased from Steraloids (Newport, RI, USA): androst-4-ene-3,17-dione, testosterone, androst-4-ene-17α-ol-3-one (epitestosterone), 5α-androstan-3α-ol-17-one (androsterone), 5β-androstan-3α-ol-17-one (etiocholanolone), 5α-androstane-3β-ol-17-one (epiandrosterone), 5α-androstan-3α,17β-diol, 5β-androstane-3α,17β-diol, 5α-androstan-3β,17β-diol, androst-4-ene-6β-ol-3,17-dione, androst-4-ene-6α-ol-3,17-dione, estr-4-ene-3,17-dione (norandrostenedione),

In vitro metabolism of androstenedione

Metabolic profiles of androstenedione after 4 h incubations in human hepatocytes from the first donor are presented in Fig. 2. Characteristic [M-15]⁺ and molecular ions of reduced metabolites of androstenedione (C₁₉O₂) were selected to build chromatograms ([M]^+· at m/z 430, 432, 434, 436, and [M-15]⁺ at m/z 415, 417, 419 and 421). Androsterone, epiandrosterone, testosterone and androstenedione were present in free fraction whereas androsterone, etiocholanolone, epitestosterone, androstenedione

In vitro vs. in vivo metabolism of androstenedione

In vivo, the oral administration of androstenedione was reported to cause important increases of the basal concentrations of androsterone, etiocholanolone and testosterone in the glucuronide fraction [13], [14]. Abnormally high urinary concentrations of androgens and a testosterone glucuronide over epitestosterone glucuronide ratio (TG/EG) above the cut-off of 6 are the probes commonly used by doping control laboratories to detect the misuse of androstenedione [1], [13], [14]. In both human

References (26)

Comité International Olympique, IOC Medical Code and Explanatory Documents,...
A. Noret, Le Dopage, Editions Vigot, Paris,...
C. Ayotte et al.
J. Chromatogr. B
(1996)
K.J. De Cock et al.
J. Pharm. Biomed. Anal.
(2001)
W. Schanzer et al.
J. Steroid Biochem. Mol. Biol.
(1996)
S.A. Wrighton et al.
Drug Metab. Rev.
(1993)
A.D. Rodrigues
Biochem. Pharmacol.
(1994)
S. Ekins et al.
A.S. Clouet et al.
Analyst
(1998)
N.G. Coldham et al.
Analyst
(1998)

A.S. Clouet-Dumas et al.

J. Steroid Biochem. Mol. Biol.

(2000)

S. Rendic et al.

J. Chromatogr. B

(1999)

J.-F. Lévesque, C. Ayotte in: W. Schanzer, H. Geyer, A. Gotzmann, U. Marek-Engelke (Eds.), Proceedings of the 17th...

Cited by (31)

Investigation of the metabolites of the HIF stabilizer FG-4592 (roxadustat) in five different in vitro models and in a human doping control sample using high resolution mass spectrometry
2017, Journal of Pharmaceutical and Biomedical Analysis
Citation Excerpt :
However, information on the metabolic fate of FG-4592 is currently lacking [8]. Investigation of the metabolite profile in vitro can be performed by using different models, such as human liver microsomes [10], human hepatocytes [11,12], equine liver microsomes and the S9 fraction [13,14]. Furthermore, it has previously been shown for other drugs that fungi of the Cunninghamella genus [9,15,16] are capable of producing the same metabolites as humans, dogs, and horses [16].
FG-4592 is a hypoxia-inducible factor (HIF) stabilizer, which can increase the number of red blood cells in the body. It has not been approved by regulatory authorities, but is available for purchase on the Internet. Due to its ability to improve the oxygen transportation mechanism in the body, FG-4592 is of interest for doping control laboratories, but prior to this study, little information about its metabolism was available. In this study, the metabolism of FG-4592 was investigated in a human doping control sample and in five in vitro models: human hepatocytes and liver microsomes, equine liver microsomes and S9 fraction and the fungus Cunninghamella elegans. By using liquid chromatography coupled to a Q-TOF mass spectrometer operated in MS^E and MSMS modes, twelve different metabolites were observed for FG-4592. One monohydroxylated metabolite was detected in both the human and equine liver microsome incubations. For the fungus Cunninghamella elegans eleven different metabolites were observed of which the identical monohydroxylated metabolite had the highest response. This rich metabolic profile and the higher levels of metabolites produced by Cunninghamella elegans demonstrates its usefulness as a metabolite producing medium. In the doping control urine sample, one metabolite, which was the result of a direct glucuronidation, was observed. No metabolites were detected in neither the human hepatocyte nor in the equine liver S9 fraction incubates.
Metabolic studies of prostanozol with the uPA-SCID chimeric mouse model and human liver microsomes
2016, Steroids
Citation Excerpt :
To overcome this problem there is a quest for alternatives to human administration for metabolism studies. Examples of such alternatives are liver fractions [12–16], hepatocyte cell cultures [17,18], chimeric mice with humanized livers [19–21] and primates [22,23]. These in vivo and in vitro model systems all have advantages as well as disadvantages.
Anabolic androgenic steroids are prohibited by the World Anti-Doping Agency because of their adverse health and performance enhancing effects. Effective control of their misuse by detection in urine requires knowledge about their metabolism. In case of designer steroids, ethical objections limit the use of human volunteers to perform excretion studies. Therefore the suitability of alternative models needs to be investigated.
In this study pooled human liver microsomes (HLM) and an uPA^+/+-SCID chimeric mouse model were used to examine the metabolism of the designer steroid prostanozol as a reference standard. Metabolites were detected by GC–MS (full scan) and LC–MS/MS (precursor ion scan).
In total twenty-four prostanozol metabolites were detected with the in vitro and in vivo metabolism studies, which could be grouped into two broad classes, those with a 17-hydroxy- and those with a 17-keto-substituent. Major first phase metabolic sites were tentatively identified as C-3′; C-4 and C-16. Moreover, 3′- and 16β-hydroxy-17-ketoprostanozol could be unequivocally identified, since authentic reference material was available, in both models.
Comparison with published data from humans showed a good correlation, except for phase II metabolism. As metabolites were in contrast to the human studies predominantly present in the free fraction. Two types of metabolites ((di)hydroxylated prostanozol metabolites) that have not been described before could be confirmed in a real positive doping control sample. Hence, the results provide further evidence for the applicability of chimeric mice and HLM to perform metabolism studies of designer steroids.
Ultra high performance liquid chromatography tandem mass spectrometric detection of glucuronides resistant to enzymatic hydrolysis: Implications to doping control analysis
2015, Analytica Chimica Acta
Citation Excerpt :
The increase observed in their concentration after T administration (around two fold at 10 h after a single oral dose of T undecanoate administration) advises for the evaluation of these metabolites as suitable markers for T misuse. Additionally, since 6OH-Andros and 6OH-Etio have been reported as metabolites of other endogenous steroids such as androstanedione [23], the screening for these glucuronides can be also useful for the detection of the misuse of a wider range of endogenous steroids. In order to confirm this potential, further research on their pharmacokinetics properties after the administration of different endogenous steroids should be performed.
Controversial results have been reported in the literature regarding the behavior of two testosterone (T) metabolites (3α-glucuronide-6β-hydroxyandrosterone and 3α-glucuronide-6β-hydroxyetiocholanolone) excreted after T administration. Due to their potential as biomarkers of T misuse, a UHPLC–MS/MS method for the direct quantification of these glucuronides was developed and validated. In addition, the main phase II metabolites of T that compose the steroid profile used for doping control purposes (glucuronides of T, epitestosterone, androsterone and etiocholanolone) were included. The method was found to be linear and with suitable LODs and LOQs for all metabolites. The average accuracies were between 86% and 120%, the RSDs for the intra- and inter-day precision were below 15% and 25% respectively. The method showed low matrix effect. Samples obtained before and after the administration of T were analyzed by both the developed UHPLC–MS/MS method and the GC–MS/MS method currently used by anti-doping laboratories. Relevant disagreements between the results obtained for 3α-glucuronide-6β-hydroxyandrosterone and 3α-glucuronide-6β-hydroxyetiocholanolone quantitation were observed. These markers seemed to be more suitable for the screening of T misuse when detected by UHPLC–MS/MS. These discrepancies were further investigated in 50 urine samples from healthy volunteers. The two methods gave highly correlated results for all metabolites that are currently included in the athlete's steroid profile confirming the reliability of the UHPLC–MS/MS method. However, the quantification of 3α-glucuronide-6β-hydroxyandrosterone and 3α-glucuronide-6β-hydroxyetiocholanolone, was only possible by using the UHPLC–MS/MS method since three interfering compounds were observed when performing the GC–MS/MS analysis with the most intense ion transitions. These results confirm the potential of the resistant glucuronides as biomarkers of T misuse. Additionally, they suggest that previously reported reference ranges for these metabolites should be reevaluated.
Characterization of desoxymethyltestosterone main urinary metabolite produced from cultures of human fresh hepatocytes
2012, Steroids
Desoxymethyltestosterone (DMT; 17β-hydroxy-17α-methyl-5α-androst-2-ene) is a designer steroid present in hormonal supplements distributed illegally as such or in combination with other steroids, for self-administration. It figures on the list of substances prohibited in sports and its detection in athlete’s urine samples is based upon the presence of the parent compound or the main urinary metabolite, which has not been characterized yet. Following its isolation from cultures of human fresh hepatocytes and S9 fractions of liver homogenates, we were able to identify this metabolite as being 17α-methyl-2β,3α,17β-trihydroxy-5α-androstane. Other minor metabolites were also characterized. The production, isolation, NMR, mass spectral analyses and chemical synthesis are presented.
Screening of 4-androstenedione misuse in cattle by LC-MS/MS profiling of glucuronide and sulfate steroids in urine
2011, Talanta
Citation Excerpt :
At that time their structure has not been elucidated, but several hypotheses (Table 1) can be suggested. Indeed, the glucuronide metabolites at m/z 481 could be associated with a hydroxy-androstan-3/17-ol-3/17-one (hydroxy-etiocholanolone, -androsterone, -epiandrosterone, -dihydrotestosterone) as described in the literature [44–47]. Concerning the sulfo-conjugate at m/z 401, an isomer of a dihydroxy-androstan-3/17-ol-3/17-one could be a possibility.
The use of anabolic agents in food producing animals is prohibited within the European Union since 1988. The illegal use of natural steroid hormones control is however still a current challenge, especially regarding the limitations of existing screening methods. In this context, the present study aimed to develop a new screening approach based on the emerging ‘untargeted profiling’ concept, but with a special emphasis on steroids phase II conjugated metabolites, in the scope of revealing potential biomarkers signing a fraudulent administration of 4-androstenedione. After extraction and separation of the urinary glucuronide and sulfate steroid fractions, each one was analyzed separately by UPLC–MS/MS using the precursor ion scan acquisition mode. This approach was carried out in order to monitor product ion characteristic of sulfate (m/z 97) and glucuronide (m/z 113) functional groups, and then to fish for any potential conjugated steroid leading to these ionic species after fragmentation. After statistical analysis, 86 metabolites (33 from steroid compounds and 53 from other unknown substances) were highlighted as potential biomarkers of 4-androstenedione abuse. After application of several robustness criteria, 26 metabolites (whom 5 were unambiguously structurally identified), were finally selected to build a statistical model which could be used as new diagnostic tool for screening purposes.
Population based evaluation of a multi-parametric steroid profiling on administered endogenous steroids in single low dose
2010, Steroids
Steroid profiling provides valuable information to detect doping with endogenous steroids. Apart from the traditionally monitored steroids, minor metabolites can play an important role to increase the specificity and efficiency of current detection methods. The applicability of several minor steroid metabolites was tested on administration studies with low doses of oral testosterone (T), T gel, dihydrotestosterone (DHT) gel and oral dehydroepiandrosterone (DHEA). The collected data for all monitored parameters were evaluated with the respective population based reference ranges.
Besides the traditional markers T/E, T and DHT, minor metabolites 4-OH-Adion and 6α-OH-Adion were found as most sensitive metabolites to detect oral T administration. The most sensitive metabolites for the detection of DHEA were identified as 16α-OH-DHEA and 7β-OH-DHEA but longest detection up to three days (after oral administration of 50 mg) was obtained with non-specific 5β-steroids and its ratios. Steroids applied as a gel had longer effects on the metabolism but were generally not detectable with universal decision criteria.
It can be concluded that population based reference ranges show limited overall performance in detecting misuse of small doses of natural androgens. Although some minor metabolites provide additional information for the oral testosterone and DHEA formulations, the topical administered steroids could not be detected for all volunteers using universal reference limits. Application of other population based threshold limits did not lead to longer detection times.

View all citing articles on Scopus

View full text

Evaluation of human hepatocyte incubation as a new tool for metabolism study of androstenedione and norandrostenedione in a doping control perspective

Abstract

Introduction

Section snippets

Standards and chemicals

In vitro metabolism of androstenedione

In vitro vs. in vivo metabolism of androstenedione

J. Chromatogr. B

J. Pharm. Biomed. Anal.

J. Steroid Biochem. Mol. Biol.

Drug Metab. Rev.

Biochem. Pharmacol.

Analyst

Analyst

J. Steroid Biochem. Mol. Biol.

J. Chromatogr. B