Factors affecting urinary excretion of testosterone

Research articleDrug Testing

and Analysis

Received: 10 June 2014 Revised: 9 March 2015 Accepted: 11 March 2015 Published online in Wiley Online Library: 28 April 2015

(www.drugtestinganalysis.com) DOI 10.1002/dta.1801

110

Factors affecting urinary excretion oftestosterone metabolites conjugatedwith cysteineAndreu Fabregat,a Josep Marcos,a,b Jordi Segura,a,b Rosa Venturaa,b

and Oscar J. Pozoa*

The implementation of the athlete steroidal passport in doping control analysis aims to detect intra-individual changes in the ste-roid profile related to the abuse of anabolic steroids. In this context, the study of intrinsic variations associatedwith eachmarker isof utmost importance. In the present work, the influence of several factors in the excretion of the recently reported testosteronemetabolites conjugated with cysteine (Δ1-AED; 1,4-androstadien-3,17-dione, Δ6-AED; 4,6-androstadien-3,17-dione, Δ6-T; 4,6-androstadien-17β-ol-3-one, and Δ15-AD; 15-androsten-3,17-dione) is evaluated for the first time. Degradation experiments at37 °C proved that, although the cysteinyl moiety is released, the variation for urinary Δ1-AED/Δ6-AED, Δ1-AED/Δ6-T ratios is lessthan 30%. Moreover, freeze/thaw cycle testing resulted in RSDs values below 15% for all the analytes. Regarding infradian vari-ability, moderate variations (below 40%) were observed. Additionally, notable alterations in the excretion of these compoundshave been observed in the earliest stages of pregnancy. UGT2B17 polymorphism, responsible for the low T/E ratio found in somepopulation, does not influence the excretion of cysteinyl compounds whereas the intake of exogenous substances (alcohol or5α-reductase inhibitors) dramatically affects their excretion. The urinary concentrations of Δ1-AED, Δ6-AED, and Δ15-AD decreased(<50%) after the ethanol intake, whereas after the administration of dutasteride, an important increasewas observed for the con-centrations of Δ6-AED, Δ6-T and Δ15-AD. Overall, the presented data describes the stability of the urinary cysteinyl steroids underthe influence ofmany factors, proving their potential as suitable parameters to be included in the steroidal module of the athlete′sbiological passport. Copyright © 2015 John Wiley & Sons, Ltd.

Keywords: steroid profile; cysteinyl metabolites; exogenous influencing factors; endogenous influencing factors; doping

* Correspondence to: Óscar J. Pozo, Bioanalysis Research Group, IMIM, Hospital delMar, Doctor Aiguader 88, 08003 Barcelona, Spain.E-mail: [email protected]

a Bioanalysis Research Group, IMIM, Hospital del Mar, Doctor Aiguader 88, 08003Barcelona, Spain

b Department of Experimental and Health Sciencies, Universitat Pompeu Fabra,Doctor Aiguader 88, 08003 Barcelona, Spain

Introduction

Anabolic androgenic steroids (AAS) are the most frequently re-ported substances by doping control laboratories. In 55.5% of thecases an atypical value for the testosterone marker was found.[1]

However, most of them did not lead to adverse analytical finding.Since testosterone (T) is naturally produced in the body, the meredetection of T is not sufficient to report an adverse analyticalfinding. Thus, anti-doping control laboratories have to face the chal-lenge of differentiating between endogenous production and ex-ogenous administration.The screening of Tmisuse remained impossible until Donike et al.

introduced the ratio between T and its 17α-epimer epitestosterone(E) as an indication of T intake.[2] Since then, the use of the T/E ratiohas been the primary marker to distinguish between endogenousconcentrations and those obtained after the exogenous administra-tion of T. A sample is considered suspicious if an abnormal value ofT/E, above the reference limit established by theWorld Anti-DopingAgency (WADA), is detected.[3] However, for the confirmation theuse of gas chromatography coupled to combustion isotope ratiomass spectrometry (GC-C-IRMS) is mandatory in order to discernbetween exogenously administered steroids to those producednaturally in the body by their relative content of carbon isotopes.[4–6]

In January 2014, a step forward was taken in the detection of Tintake with the incorporation of the steroid profile in the Athlete Bi-ological Passport (ABP).[3] The ABP allows for the continuous andsystematic storage and evaluation of the athlete’s parameters of

Drug Test. Analysis 2016, 8, 110–119

the steroid profile in a longitudinal way. This approach uses individ-ual reference limits for each athlete based on the data obtainedthroughout all the tests conducted worldwide.[7] Owing to these in-dividual reference limits, the ABP is sensitive to detect endogenoussteroid administrations by small but noticeable changes in themarkers of the steroid profile. However, this characteristic alsomakes ABP sensitive to any source of intra-individual variabilityassociated to the markers. For this reason, studies of the differentfactors that can alter the steroid profile which in turn may lead toa false suspicion of doping are of utmost importance.[8,9]

Currently, the steroid profile includes the quantification of thefollowing compounds: T, E, Androsterone (Andros), Etiocholanolone(Etio), 5α-androstane-3α,17β-diol (5α-Adiol), and 5β-androstane-3α,17β-diol (5β-Adiol).[3] The effect of different factors on the ex-creted concentrations of these compounds has been largely stud-ied (see sections Sample preparation, Endogenous factors,Exogenous factors).

Copyright © 2015 John Wiley & Sons, Ltd.

Factors affecting cysteinyl testosterone metabolites

Drug Testing

and Analysis

The inclusion of the excreted concentrations and ratios of otherminor T metabolites in the steroid profile has been presented as apowerful approach in order to increase the diagnostic specificityof ABP.[10–12] Before including these metabolites in the ABP, it isnecessary to know how they are altered by common factors affect-ing the steroid profile.

Several T metabolites released after an alkaline treatmentof the urine sample (1,4-androstadien-3,17-dione (Δ1-AED), 4,6-androstadien-3,17-dione (Δ6-AED), 4,6-androstadien-17β-ol-3-one(Δ6-T) and 15-androsten-3,17-dione (Δ15-AD)) have been reportedand characterized.[13] These metabolites were shown to originatefrom a phase II cysteinyl conjugate.[14,15] For Δ15-AD no confirma-tion of the conjugate with cysteine was possible due to the ab-sence of a commercial standard. However, since this compoundhas been only observed after the alkaline treatment of the urine,by analogy with Δ1-AED, Δ6-AED and Δ6-T, it is expected that Δ15-AD come from a conjugate with cysteine.[13] A method for thequantification of these compounds in urine was developed andvalidated. Due to the low sensitivity observed for the direct detec-tion of the cysteinyl conjugates, the method includes a cleavageof the cysteine moiety before the analysis. Due to the lack of ref-erence material for Δ15-AD, a semi quantification method wasperformed using Δ6-T as reference standard.[16]

The usefulness of these compounds for the detection in the dop-ing control field was also evaluated.[17,18] In particular, it was dem-onstrated that the use of Δ1-AED/Δ6-AED and Δ1-AED/Δ6-T ratiosas markers of T abuse improved the detection capabilities of thecommonly employed T/E ratio.[17] The use of the ratios involvingΔ15-AD (Δ1-AED/Δ15-AD, Δ6-AED/Δ15-AD and Δ6-T/Δ15-AD) showedtheir usefulness when dealing with the topical application of T andother pro-hormones such 5α-dihydrotestosterone (DHT).[18] Addi-tionally, the ratios involvingmetabolites Δ6-AED andΔ6-T exhibitedan increase after the administration of dehydroepiandrosterone(DHEA), being the ratio Δ6-T/Δ15-AD the best marker for itsdetection.[13] However, it is still unknown how the values of thesemetabolites/ratios are altered by common factors affecting the ste-roid profile.

The aim of this study is to evaluate how the concentrations andratios of the T metabolites conjugated with cysteine are affected bythe factors which are known to influence the steroid profile. Theevaluated parameters are divided in three different categories:sample preservation, endogenous factors and exogenous factors.The influence of these variables in the current steroid profile is fur-ther described below.

111

Sample preservation

Since in most of the cases, the analysis of urine samples for drugtesting is not performed immediately after their collection, it is veryimportant to evaluate the effect of storage conditions.[19]

The fact that the urine samples are collected in non-sterile condi-tions offers the micro-organisms the opportunity to grow, espe-cially when samples are stored for a long time.[9] It is known thatmicroorganisms can cause urinary alteration by hydrolysis of glucu-ronide and sulphate and can cause oxireductase reactions ofsteroids.[20–22] Therefore, bacterial activities in urine may cause sig-nificant changes in the measured steroid concentrations. The pri-mary reaction that occurs is the deconjugation of glucuronidesand sulphates.[22,23] Since in doping control tests only the combinedglucuroconjugated and free fraction are analyzed, and the influ-ence of deconjugated sulphate after bacterial hydrolysis is

Drug Test. Analysis 2016, 8, 110–119 Copyright © 2015 John W

unknown, this primary reaction might influence the measuredsteroid concentrations. In addition to that, this deconjugation isnormally associated to other transformations like uncontrolledoxidoreduction or hydrolysis reactions, whichmight lead to the al-teration of several steroid profile parameters. For instance, Marecket al. reported an increased T/E from 5.3 to 9.8, determined in acombined fraction of conjugated and unconjugated steroids.[8]

Markers of bacterial urine degradation are the accumulation of5α-androstan-3,17-dione (5α-AD) and 5β-androstan-3,17-dione(5β-AD) which are the product of Andros and Etio deconjugation re-spectively, followed by a bacterial 3-hydroxysteroid-dehydrogenaseactivity.[24]

Additionally, in anti-doping controls, the urine sample is collectedand divided into two aliquots (A and B). While aliquot A is analyzedimmediately upon arrival to the laboratory and may suffer severalfreeze/thaw cycles for conducting different tests, the B aliquot isstored at �20 °C waiting for a possible counter analysis.[25] Knowl-edge about the stability of the steroids to different freeze/thawcycles is critical for a proper interpretation of the results. In thecase of the urinary T/E ratio, a high stability has been observed.[24]

Endogenous factors

Due to the homeostasis of biosynthesis and metabolism of endog-enous steroids,[26] the amounts of steroids in 24h periods areconstant for a given individual.[26] However, the concentrations ofthe eliminated endogenous steroids may vary with the urinary flow.The use of ratios between steroid compounds minimizes thisfluctuation.[27] Despite this, T/E ratio can vary greater than 30 % inlongitudinal studies.[28,29] Besides, due to the influence of the dailycircadian rhythm in the steroid excretion, plasma variations be-tween 20–40% for T have been reported.[30] This circadian rhyth-micity in the production of T is reflected in a moderate circadianrhythm in the urinary T excretion.

Some small differences were reported between genders. For in-stance, while the intra-individual variations in the T/E ratio in malesis expected to be less than 30%,[31] the observed variations in fe-males are higher due to the menstrual cycle,[28] and also in someextent to a higher analytical imprecision derived from the lower Tand E concentrations.[32] Variation between genders were reportedin the concentrations of androgens exhibiting lower concentrationin female urine samples.[28] However, no significant changes wereobserved for the T/E ratio between genders.[31,32] Other factors likeextensive exercise can also influence the steroid excretion.[33]

Pregnancy dramatically affects thematernal excretion of steroids.The changes observed in the steroid excretion aremainly due to thefact that products of the feto-placental unit have no other exit thanthe maternal serum and urine.[34] Significant differences wereobserved in the T/E values in the first trimester of pregnancy whencomparing with basal values. Whereas T remained almost unalteredduring gestation, a significant rise was observed for E during thefirst weeks of pregnancy leading to a decrease in the T/E ratio.[35]

At present, pregnancy is the only influencing factor that has beenevaluated for the conjugated steroids with cysteine, and differentbehaviours were revealed.[35] A substantial rise of Δ1-AED and adecrease of Δ6-T concentrations were observed during the first tri-mester of pregnancy. Subsequently, all the ratios involving thesecompounds were affected. However, the Δ15-AD and the ratiosinvolving this compound were not monitored in that study.

Finally, large differences in the steroid excretion between Asianand Caucasian populations have been widely reported.[36] These

iley & Sons, Ltd. wileyonlinelibrary.com/journal/dta

A. Fabregat et al.

Drug Testing

and Analysis

112

differences are derived from a deletion in the gene UGT2B17 whichencodes the enzyme responsible for the glucuronidation process ofthe T but not for the E.[37] This deletion was found to be more com-mon in Asian populations. Since T is excreted mainly as a glucuro-nide conjugate, populations showing a deletion in this gene havelower T/E values than the rest.[38] For this reason T/E exhibited a bi-modal distribution with two subpopulations with respective meansof T/E around 0.1 and 1.0.[24,39] Consequentially, if a single cut-offvalue for the T/E ratio is applied to the whole population, athleteswith the deletionwill have an unfair advantage.[40] Therefore, the riskof a false negative result is higher in athletes carrying the UGT2B17gene deletion. The application of the ABP will drastically minimizethis problem since a subjected-based reference threshold is used.

Exogenous factors

It is known that several exogenous factors can influence the steroidpattern in urine. For instance, ethanol consumption implies a directchange in the steroid production. Several hypotheses have beenpostulated in the literature for this behaviour.[41,42] The most ac-cepted one is that ethanol metabolism increases the ratio ofNADH/NAD+, which results in a suppressed steroid oxidation.[41]

The effects of the ethanol in the T/E ratios have been extensivelystudied.42-44 The intake of acute alcohol doses has been demon-strated to raise urinary T concentration, while E levels remainunaltered.[43,45] Consequently, the oral intake of ethanol leads to asignificant increase on T/E ratio. The determination of ethyl glucuro-nide (EtG) and ethyl sulphate (EtS) are well-established markers forthe ethanol consumption in clinical toxicology.[46,47] Therefore, inorder to avoid false positive results, ethanol markers have to bemonitored in suspicious urine specimens.[48]

On the other hand, the use of 5α-reductase inhibitors such asdutasteride and finasteride can lead to important modifications inthe steroid profile. These kind of drugs are used therapeutically totreat benign prostatic hyperplasia[49] as well as male baldness[50]

by decreasing the 5α-reductase activity, and consequently thelevels of the potent endogenous androgen DHT. These drugs affectthe production of 5α-steroids such as Andros or 5α-Adiol.[51] Afterthe intake of a single 5mg dose of finasteride the Andros/Etio ratiowas reduced to a 10% of its original value in 24h, and the alterationwas still present 8 days after the intake.[51]

Experimental

Chemicals and reagents

Androsta-4,6-dien-3,17-dione (Δ6-AED) and 17β-hydroxy-androsta-4,6-dien-3-one (Δ6-T), 5α-androstanedione (5α-AD), and 5β-androstanedione (5β–AD) were obtained from Steraloids Inc.(Newport, Rhode Island, USA). Androsta-1,4-dien-3,17-dione (Δ1-AED)was purchased from NMI (Pymble, Australia). Testosterone (T),epitestosterone (E), androsterone (Andros), etiocholanolone(Etio), 5α-androstane-3α,17β-diol (5α-Adiol), 5β-androstane-3α,17β-diol (5β-Adiol), as well as the internal standardsmethandienone (MD) and methyltestosterone (MT) were obtainedfrom Sigma-Aldrich (St Louis, MO, USA). The β-glucuronidasepreparation (from Escherichia coli, type K12) was purchased fromRoche Diagnostics (Mannheim, Germany). The DRI® EtG enzymeimmunoassay was obtained from Thermo Fisher Scientific(Waltham, MA, USA).

wileyonlinelibrary.com/journal/dta Copyright © 2015 Jo

Analytical grade potassium carbonate, potassium hydroxidepellets, di-sodium hydrogen phosphate, sodium hydrogen phos-phate, tert-butyl-methyl ether, and ammonium iodide were ob-tained from Merck (Darmstadt, Germany). The derivatizationreagent N-methyl-N-trimethylsilyl-trifluoroacetamide (MSTFA) wasfrom Karl Bucher Chemische Fabrik GmbH (Waldstetten, Germany)and 2-mercaptoethanol was from Sigma-Aldrich (St Louis, MO, USA).

Acetonitrile and methanol (LC gradient grade), formic acid andammonium formate (LC/MS grade) were purchased from Merck(Darmstadt, Germany). Milli Q water was obtained using a Milli-Qpurification system (Millipore Ibérica, Barcelona, Spain).

Instrumentation

LC-MS/MS

Selected reactionmonitoring (SRM)methodwas carried out using atriple quadrupole (Quattro Premier XE) mass spectrometer equippedwith an orthogonal Z-spray-electrospray ionization source (ESI)(Waters Associates, Milford, MA, USA) interfaced to an UHPLC sys-tem, Acquity (Waters Associates) for the chromatographic separa-tion. Drying gas as well as nebulizing gas was nitrogen. Thedesolvation gas flow was set to approximately 1200 L/h and thecone gas flow to 50 L/h. A cone voltage of 25 V, and a capillaryvoltage of 3.0 kV were used in positive ionization mode. The nitro-gen desolvation temperature was set to 450 °C and the sourcetemperature to 120 °C. The LC separation was performed usingan Acquity UHPLC BEH C18 column (100mm x 2.1mm i.d.,1.7μm particle size) (Waters Associates, Milford, MA, USA), at flowrate of 300μL/min. Water and methanol both with formic acid(0.01 % v/v) and ammonium formate (1mM) were selected as mo-bile phase solvents. Additional details such as detailed gradientand SRM method used can be found elsewhere.[16]

GC-MS

The GC-MS analysis was performed on an Agilent 6890 GC systemcoupled to a 5975B Vl MSD mass spectrometer from Agilent Tech-nologies (Palo Alto, CA, USA). The instrument was equipped with a100 % dimethylpolysiloxane Ultra1 column (17m x 0.2mm i.d.,0.11μm film thickness) from J&W (Palo Alto, CA, USA). The temper-atures of the instrument parts were 280 °C for the injector, 280 °C forthe transfer line, 230 °C for the ion source and 150 °C for the quadru-pole. TheGC temperature programwas: at 0min, 180 °C; then 3 °C/minup to 230 °C, then 40 °C/min up to 310 °C, then 3min at 310 °C.

Helium was used as a carrier gas which was under constant pres-sure of 19psi. For the steroid quantification Selective Ion Monitoring(SIM) was used with a dwell time of 20ms for all monitored ions.Additionally,m/z 272 was monitored to screen for the possible pres-ence of the mono-trimethylsilyl derivatives of androsterone andetiocholanolone which is an indication of incomplete derivatization.

Sample preparation procedures to measuresteroid metabolites

Testosterone metabolites conjugated with cysteine

Due to the low sensitivity achieved for the direct detection of thecysteinyl compounds in urine samples,14 the quantification of Δ1-AED, Δ6-AED, Δ6-T, and Δ15-AD was performed by an indirectmethod after the cleavage of the cysteine moiety based on theLC-MS/MS approach described elsewhere.[16] Briefly, after addition

hn Wiley & Sons, Ltd. Drug Test. Analysis 2016, 8, 110–119


Drug Testing

and Analysis

of 50μL of the ISTD solution (MD, 1μg/mL) 2.5mL of urine were ba-sified with 300μL of a 6M KOH solution. The mixture was heated at60 °C for 15min, followed by a liquid-liquid extraction (LLE) with6mL of tert-butylmethylether. The sample was centrifuged andthe organic layer separated and evaporated. The residue was dis-solved into 150μL of a mixture of water:acetonitrile (1:1, v/v). Fi-nally, 10μL were injected into the LC-MS/MS system.

For the microbial degradation studies (see sections on Degrada-tion Sample preservation factors), an additional LLE with 6mL tert-butylmethylether was performed prior to the alkaline treatment inorder to separate the analytes present in the free faction. Thereaf-ter, the sample was treated as stated above.

For Δ15-AD no confirmation of the conjugated with cysteine inurine was possible due to the lack of authentic reference material.However, this metabolite was only observed after the alkaline treat-ment of the urine.[13] Thus, based on an analogy with Δ1-AED, Δ6-AED and Δ6-T, it was hypothesized that Δ15-AD originates from aconjugate with cysteine. Additionally, due to lack of available refer-ence material for Δ15-AD, a semi-quantification was performedusing Δ6-T as a reference standard as described elsewhere.[16]

Conventional testosterone metabolites

According to the WADA technical document, laboratories have toreport the concentrations corresponding to the six steroidsincluded in the steroid module of the ABP. These steroids are: T,E, Andros, Etio, 5α-Adiol and 5β-Adiol, and have to be quantifiedby GC-MS after the formation of the corresponding bis-TMSderivatives.[3] In the present study, these quantifications wereperformed by using a GC-MS procedure adapted from a methoddeveloped elsewhere.[10] Fifty μL of an ISTD solution (MT at2μg/mL) and 1mL of phosphate buffer (1M, pH7) were addedto 2.5mL of urine. The mixture was hydrolyzed with 50μL ofβ-glucuronidase during 1h at 55 °C. After cooling, the samplewas basified with 250μL of K2CO3 (25%, w/v). A LLE was performedby addition of 6mL of tert-butylmethylether. The tubes werecentrifuged, the organic layer separated, and evaporated undera nitrogen stream. The dry residue was derivatized with 50μL ofa MSTFA/NH4I/2-mercaptoethanol (1000:2:6, v/w/v) mixture for20min at 60 °C. Finally, 1μL was injected into the GC-MS systemworking in split mode (20:1).

Ethyl glucuronide detection

For the detection of Ethyl glucuronide (EtG) the DRI® EtG enzymeimmunoassay was used. This immunoassay provides a semi-quantitative determination of EtG in human urine. The sampleprocedure for the detection of EtG was based on the methoddescribed elsewhere.[52]

Correction of the urine dilution

In order to reduce the variability due to urine dilution, the urinaryconcentrations were corrected to a specific gravity of 1.020 accord-ing to WADA’s technical document on endogenous steroids.[53]

113

Studies on influencing factors

Degradation

Urine samples from 3 male and 3 female volunteers (age 39±11 years; weight, 72±14 kg)were collected in non-sterile containers.


Each urine sample was divided into 6 aliquots of 5mL each for thesubsequent incubation at 37 °C. Aliquots were incubated during 0,2, 3, 4 and 7 days, respectively. Every selected day, the aliquot wasdivided into two different sub-aliquots of 2.5mL and analyzed.

Freeze/thaw stability

The effect of different freeze/thaw cycles were studied using urinesamples from 3 male and 3 female volunteers (age 32±7 years;weight, 73± 12 kg). For this purpose, 4 aliquots of each urine sam-ple collected underwent 0, 1, 2, or 3 freeze (at�20 °C) and thawing(at room temperature) consecutive cycles, respectively.

Infradian variations and circadian variations

For the study of infradian variations, urine samples from 8male and8 female volunteers (age 39±10years; weight, 65±16kg) were col-lected weekly during two consecutive months. Once collected,samples were kept at �20 °C until analysis. In order to minimizethe long-term component of the analytical variance, samples wereanalyzed in two consecutive analytical batches.

Regarding the circadian variations, urine samples from the same16 volunteers were collected during one day at four different times:9 a.m., 1 p.m., 9 p.m., and 12p.m.

Gender variations

For the gender variations, the first morning urine from 8 males and8 female volunteers (age 39±10 years; weight, 65± 16 kg) wereanalyzed.

Ethnic/ genotype variations

Samples were selected in order to investigate if the studiedmarkerswere affected by a deletion polymorphism in the gene coding theglucuronosyltransferase responsible for the T glucuronidation(UGT2B17) which is present in 81% of the Asian population, andleads to low urinary T/E values.[38] Three urine samples from 6 Asianvolunteers (age 51±7 years; weight 75±12 kg; T/E values below 0.2)and 6 Caucasian volunteers (age 27±5 years; weight 71±10kg; T/Ebetween 1 and 3) were investigated.

Pregnancy

The effect of pregnancywas evaluated by analyzing spot urine sam-ples from a pregnant woman (age 35 years; weight, 45 kg) collectedweekly before (n= 9) pregnancy and during (n= 8) the first trimes-ter of gestation.

Ethanol consumption

For the ethanol (EtOH) consumption studies, 18 male volunteers(age 27±5 years and weight 75±5 kg) were selected. Three differ-ent studies (Placebo, sample dose 24g of EtOH and sample dose of42g EtOH) were conducted. The ethanol studies were performedusing a pharmaceutical preparation of absolute EtOH diluted withlemon flavored water (200mL) at the required concentration. Forthe placebo study 200mL of water with lemon flavour was givento the volunteers. The patients were randomized to treatment withplacebo, 24 g of EtOH or 42g EtOH (six volunteers for eachtreatment).


A. Fabregat et al.

Drug Testing

and Analysis

114

Different urine samples were collected one sample before thetreatment (0h) and samples at 2, 4, 6 12, and 24h after theadministration.

5α-reductase inhibitor studies

For the 5α-reductase inhibitor studies, a single dose of 0.5mg ofdutasteride (one capsule of Avodart®) was given to onemale volun-teer (aged 49 years; weight 87 kg). Urine samples before the admin-istration (0h) and at 3, 9, 15, 39, and 72h after the administrationwere collected.

Ethical approval

Ethical approvals for all these studies was granted by the clinical re-search ethical committee of our institute (CEIC-Parc de Salut Mar,Barcelona, Spain). All subjects participating in this study gave theirwritten informed consent.

Statistical analysis

Statistical analyses were performed using SPSS version 18 (30 July2009). For the statistical analysis, a t-Student and its non-parametricanalogous Mann–Whitney U tests were used. The normality of thedifferent markers evaluated in this study was evaluated by using aKolmogorov-Smirnov test. Statistical significance was taken asp< 0.05.

Results and discussion

Sample preservation factors

Microbial degradation

The microbial degradation of the urine sample was evaluated bymonitoring the concentration of the commonly used markers forsample degradation, i.e., 5α-AD and 5β-AD.[54] As expected, the con-centrations of these steroids increased with the incubation time of

Figure 1. Representation along the time of the concentration (μg/mL) of the cvariation of the hydrolyzed fraction, % of free fraction and % of total amount foconjugated with cysteine (d) Δ1-AED, (e) Δ6-AED, (f) Δ6-T, (g) Δ15-AD (h) Δ1-AED/Δ6-


the urine sample at 37 °C (Figures 1a and 1b). Whereas the concen-trations of 5α-AD showed a rapid increase exhibiting amaximum atthe third day of incubation (around 5μg/mL), the concentrations of5β-AD exhibited a progressive increase, reaching the value of5μg/mL at the seventh and last day of incubation (Figure 1b).The calculated concentrations of these compounds exceeded byfar the values proposed as an indication limit for degradation(10 ng/mL for 5α-AD and 20ng/mL 5β-AD) even in the first col-lected urine (after 2 days of incubation at 37 °C).[54] The maximumvalues observed for these markers exceeded more than 100 foldthese values (Figures 1a and 1b) proving the intense microbial ac-tion in the incubated samples.

The results observed for the T/E ratio (Figure 1c) in this study weresimilar to those found in a previous study, where no significantvariations (less than 30%) in the T/E values were observed.[55,56]

The effect of the microbial degradation on the selected cysteinylsteroids was evaluated by calculating the percentage of the freefraction and the percentage of the conjugated fraction in eachvolunteer. The mean and standard deviation (SD) for all analyzedvolunteers were represented along the days of incubation. Addi-tionally, the total amount which is the sum of the two analyzed frac-tions (conjugated and free) was monitored (Figure 1).

A similar behaviour after different days of incubation at 37 °Cwasobserved for the urinary concentrations of Δ1-AED, Δ6-AED, Δ6-T,and Δ15-AD. An important increase in the percentage of the freecompound in harmony with a decrease of the percentage of theconjugated compound was observed (Figures 1d–1g). For Δ15-AD,a higher variation in the percentage of the conjugated form andthe percentage of the free formwas observed during the incubation(Figure 1g). These results are in agreement with the release of thecysteine moiety. A similar release behaviour was observed for Tglucuronide after the incubation for 72 h at 37 °C 20,24. Since thecleavage of the cysteine moiety is the first step of the analyticalmethod (see section Testosterone metabolites conjugated withcysteine), the fact that the degradation process of these compoundsmainly leads to the cleavage of the cysteine conjugates minimizesits influence on the quantification of these steroids.

ommon markers of microbial degradation (a) 5α-AD and (b) 5β-AD, the % ofr the common marker (c) T/E and the selected markers among compoundsAED, (i) Δ1-AED/Δ6-T, (j) Δ1-AED/Δ15-AD, (k) Δ6-AED/Δ15-AD and (l) Δ6-T/Δ15-AD.



Drug Testing

and Analysis

According to the quantitative results, the ratios Δ1-AED/Δ6-AED,Δ1-AED/Δ6-T showed variations less than 30% during the incuba-tion experiments (Figures 1h and 1i). These results are similar thanthose observed for the T/E ratio. However, due to the large varia-tions observed for Δ15-AD, higher variations (from 40 % to 300 %)were observed for the ratios involving Δ15-AD (Figures 1j and 1l).

Freeze/thaw cycles

The stability of the cysteinyl compounds and the ratios betweenthem during freeze/thaw conditions were evaluated by calculatingthe individual relative standard deviation (RSD) for each markerafter 3 freeze/thaw cycles in samples from 6 volunteers.

All compounds and tested ratios showed RSDs values below 15%after the different freeze/thaw cycles applied to the samples(Table 1). These results are comparable to those previously describedfor the T/E ratio, where variations less than 10 % was observedafter 3 cycles.[55]

Table 2. Effect of gender and genetic variations on the urinary values

Endogenous origin factors

Infradian variations

The infradian variations of the steroids conjugated with cysteine,and the ratios among them, were evaluated by calculating the indi-vidual RSDs values for each marker in weekly collected samples(8 consecutive weeks) for 8 male and 8 female volunteers. Then,the mean of the RSDs obtained after the analyses of the 6 samplesfor each volunteer were calculated and the results stratified bygender are summarized in Table 1.

Variations less than 36% were usually observed for the markersΔ1-AED, Δ6-AED, Δ6-T and Δ15-AD in male volunteers (Table 1).Regarding female samples, higher RSDs values (between 33%and 40%) were observed for the urinary concentrations of thesecysteinyl compounds (Table 1). This fact can be attributed to hor-monal changes derived from the menstrual cycle as it was reportedfor the excretion of androgens in female samples 33. The use of theΔ1-AED /Δ6-T and Δ1-AED/Δ6-AED ratios minimized the urinary var-iations for both male and female volunteers (Table 1). Nevertheless,the use of the ratios involving Δ15-AD showed higher RSD values.The high variability observed for these ratios is probably due tothe fact that for Δ15-AD a semi-quantification using Δ6-T as

Table 1. Effect of freeze/thaw cycles, infradian and circadian variationson themean RSD andmaximum RSD values for the cysteinyl concentra-tions and the ratios selected between these compounds

Mean RSD (Maximum RSD) in %

Freeze/thaw Infradian Circadian

Compound Volunteers Male Female Male Female

Δ1-AED 14 (20) 26 (34) 40 (58) 27 (33) 41 (63)

Δ6-AED 11 (18) 23 (32) 36 (51) 31 (21) 39 (77)

Δ6-T 3 (6) 24 (28) 33 (55) 30 (26) 42 (71)

Δ15-AD 6 (9) 36 (47) 34 (48) 27 (31) 43 (74)

Ratio

Δ1-AED /Δ6-T 12 (19) 17 (24) 24 (43) 15 (28) 18 (39)

Δ1-AED/Δ6-AED 5 (6) 18 (23) 22 (30) 15 (18) 16 (40)

Δ1-AED/Δ15-AD 12 (18) 37 (61) 32 (48) 25 (56) 32 (50)

Δ6-AED/Δ15-AD 10 (16) 35 (53) 24 (31) 31 (43) 29 (35)

Δ6-T/Δ15-AD 5 (9) 46 (57) 30 (55) 31 (43) 28 (36)


reference compound was performed due to the lack of a referencematerial.[16]

For the classical T/E marker, a maximum RSD of 55% was ob-served by Catlin et al.[28] in drug-free males for three or more sam-ples taken at monthly or greater intervals. In another study, afteranalyzing samples from 796 athletes, it was concluded that the90th percentile for the RSD of T/E was 58%.[29]

Thus, the variations observed for at least two cysteinyl markers(the Δ1-AED /Δ6-T and the Δ1-AED/Δ6-AED ratios) were in the samerange as those observed for the T/E ratio.

Circadian variations

Similarly to the infradian variations, daily variations of male and fe-male volunteers for the studiedmarkers were evaluated by calculat-ing the individual RSDs values in samples collected at differenttimes during the same day. Results are summarized in Table 1.

Variations around 30% were in general observed for the urineconcentrations ofΔ1-AED, Δ6-AED, Δ6-T andΔ15-AD for male volun-teers. Higher variations (around 40%) were reported for the urineconcentrations of these compounds in female volunteers (Table 1).Different behaviours were observed in the excretion of these com-pounds during the day depending on the gender. These data sug-gested the presence of a circadian rhythm in the excretion of thesesteroids conjugated with cysteine. The use of the ratios Δ1-AED/Δ6-T and Δ1-AED/Δ6-AED allowed minimizing the urinary variations ofthe circadian rhythm with RSDs around 15% and 20% for male andfemale volunteers, respectively (Table 1). In agreement with thoseresults observed for the infradian variations, higher RSDs were ob-served for those ratios involving Δ15-AD.

Daily circadian variations observed for these compounds weresimilar to those results observed for the excretion of androgensboth in male[31] and female[32] during the day.

Gender variations

Genders differences were evaluated by calculating the mean andSD of male and female samples, respectively (Table 2).

of the cysteinyl compounds and the ratios selected between thesecompounds

Mean ± SD

Gender variations Genetic variations

Compound Male Female Caucasian Asian

Δ1-AED (ng/mL) 0.26 ± 0.11 0.31 ± 0.37 0.24 ± 0.17 0.24 ± 0.14

Δ6-AED (ng/mL) 16.8 ± 10.1** 25.7 ± 17.6** 19.6 ± 14.2 20.6 ± 17

Δ6-T (ng/mL) 3.7 ± 1.9* 4.3 ± 2.9* 3.3 ± 2.3 2.6 ± 2.2

Δ15-AD (ng/mL) 1.1 ± 0.78** 0.58 ± 0.57** 1.6 ± 1.3 1.5 ± 1.3

Ratio

Δ1-AED /Δ6-T 0.96 ± 0.60 0.78 ± 0.57 0.9 ± 0.67 1.2 ± 0.9

Δ1-AED/Δ6-AED 1.8 ± 0.77 1.2 ± 0.76 0.17 ± 0.03 0.13 ± 0.02

Δ1-AED/Δ15-AD 3.2 ± 1.8** 6.3 ± 5.1** 2.3 ± 4.1 3.1 ± 4.5

Δ6-AED/Δ15-AD 2.1 ± 1.5** 5.4 ± 2.7** 1.7 ± 1.6 1.7 ± 1.2

Δ6-T/Δ15-AD 4.7 ± 3.6** 9.6 ± 5.6** 2.9 ± 2.3 2.2 ± 1.5

* p< 0.05

** p< 0.01


115

A. Fabregat et al.

Drug Testing

and Analysis

116

The concentrations of Δ6-AED, Δ6-T and Δ15-AD showed signifi-cant differences (t-test; p< 0.01, p< 0.05 and p< 0.01, respec-tively) between genders (Table 2). Whereas higher values werefound in female samples for Δ6-AED (25.7 ng/mL vs. 16.8ng/mL inmale samples) and Δ6-T (4.3 ng/mL vs. 3.7ng/mL in male samples),the urinary concentrations of Δ15-AD were shown to be higher inmale samples (1.1 ng/mL vs 0.58 ng/mL in female samples). In con-trast, no significant differences were observed for the urinary con-centrations of Δ1-AED (Table 2). Results for Δ6-AED and Δ6-T arenot in agreement with the observed for the excretion of androgenswhere lower concentrations are generally observed in femalesamples.[57]

Regarding the selected ratios among these compounds, no sig-nificant differences were observed for the ratios Δ1-AED/Δ6-T andΔ1-AED/Δ6-AED (Table 1). This is in agreement with results ob-served for the T/E ratio, where no differences between gendershave been observed.[31,32] However, due to the large variations ob-served in the concentrations of Δ15-AD between genders, notabledifferences were observed when using those ratios involving thiscompound (Table 2).

Genetic variations

For the study of potential genetic variations the mean and SDvalues calculated for each marker were calculated for both Asian(T/E below 0.2) and Caucasian (T/E between 1 and 3) volunteers.The possible differences between these two ethnics groups werestudied by using a t-student test with a p< 0.05. The data observedfor these two groups are summarized in Table 2.The expected difference between Asian and Caucasian volun-

teers for the T/E ratio was confirmed: values of 0.16± 0.05 and 1.7± 0.97 were obtained for the T/E ratio in Asian and Caucasian volun-teers, respectively.Moreover, no significant differences were observed for the con-

centrations of Δ1-AED, Δ6-AED, Δ6-T and Δ15-AD among the twogroups (Table 2). Similarly, no significant differences were observedwhen applying the selected ratios among cysteinyl steroids(Table 2). This fact contrasts with the results reported for the T/Emarker in this study and previous works.[37,38] Thus, these datashows that the polymorphism in the UGT2B17 (observed in the81% of the Asian population)[39] has no effect on the excretion ofthe cysteinyl metabolites. This behaviour for the cysteinyl metabo-lites was expected since UGT2B17 is related to the glucuronidationprocess.[37,38]

Pregnancy

In order to evaluate the effect of pregnancy in the excretion ofcysteinyl markers during the first stages of gestation, the meanand SD of the quantification of the basal samples and samples

Table 3. Effect of pregnancy: Comparison of concentrations in basal sampmetabolites Δ1-AED, Δ6-AED, Δ6-T and Δ15-AD, and selected ratios of these co

Δ1-AED Δ6-AED Δ6-T Δ15-AD T/E Δ1-AED

Basal 0.21 ± 0.08 36 ± 11 4.1 ± 0.5 0.1 ± 0.05 0.03 ± 0.01 0.51 ±

1st 0.40 ± 0.16 40 ± 12 3.9 ± 0.88 1.0 ± 0.4 0.01 ± 0.003 1.0 ±

p 0.018* 0.480 0.865 <0.001** 0.006** <0.00

* p< 0.05

** p< 0.01


collected during the first trimester of pregnancy was calculated(Table 3). The differences were compared by using a t-student test.

The urinary amounts of Δ15-AD exhibited a significance increase(around 10 times) in the first trimester of the pregnancy as can beseen in Table 3. Due to the large increases observed for Δ15-AD,the ratios involving this compound showed a significant variationin their values during the first weeks of pregnancy. In contrast, sig-nificant increases in the ratios Δ1-AED/Δ6-T and Δ1-AED/Δ6-AEDwere previously reported (for more details see Fabregat et al.[35]).

The behaviour observed for the ratios involving Δ15-AD followssomehow the results observed for the T/E ratio, where a decreaseduring the first trimester of pregnancy was observed (Table 3). Theeffect is however more pronounced in the cysteinyl compounds.

Exogenous origin factors

Ethanol

The effect of the EtOH consumption on the selected markers wasevaluated by calculating the variations observed after the ingestionof placebo, 24g and 42g of EtOH. For this purpose, the relative var-iations of the markers after the administration were calculated.Figure 2 represents themean and SD for all analyzed volunteers, to-gether with the concentrations of ethylglucuronide (EtG) in all urinesamples.

As expected, a dose dependent increase in the EtG concentra-tions was observed after the EtOH ingestion (Figure 2a). In the sameway, T/E values showed a two fold increase after the intake of 42gof EtOH (Figure 2b). Basal values recovered 12h post-administration.These results are in agreement with previously published studieson the effect of the EtOH on the steroid profile.[42,44]

The variations in concentration of the cysteine conjugated ste-roids and the ratios between them after the ingestion of EtOH aredepicted in Figures 2c-2k.

Similar to the results observed for the circadian rhythm (see sec-tion on Circadian variations), placebo samples showed variationsless than 40% for all studied concentrations and ratios withincysteinyl compounds (data not shown).

As it can be observed in Figures 2c–2f, no substantial dose de-pendent differences were observed for the concentrations of Δ1-AED, Δ6-AED, Δ6-T and Δ15-AD. However different behaviours wereobserved for the urinary concentrations of these compounds afterthe ingestion of EtOH. The urinary concentrations of Δ1-AED, Δ6-AED, and Δ15-AD exhibited a decrease in their concentrations (lessthan 50% of basal values) following EtOH administration; thereafterthese compounds recovered the normal values (Figures 2c, 2d, and2f, respectively). However, the urinary concentrations of Δ6-Tshowed a slight decrease in the first hours of the administrationand, thereafter, an increase of around 100% was produced(Figure 2e). The decrease observed for Δ1-AED, Δ6-AED and Δ15-AD

les and in samples obtained during the first trimester of pregnancy formpounds as reported[35] and the ratios involving Δ15-AD

/ Δ6-T Δ1-AED/ Δ6-AED Δ1-AED/Δ15-AD Δ6-AED/Δ15-AD Δ6-T/Δ15-AD

0.09 0.57 ± 0.08 11.2 ± 7.5 20.6 ± 15.4 28.5 ± 21.01

0.46 0.97 ± 0.25 3.9 ± 1.4 4.0 ± 0.88 3.8 ± 0.4

1** <0.001** 0.019* 0.009** 0.005*


Figure 2. Variations observed for the marker for EtOH ingestion (a) EtG; and the variations observed for the ratios (b) T/E, (c) Δ1-AED, (d) Δ6-AED, (e) Δ6-T,(f) Δ15-AED (g) Δ1-AED/Δ6-AED, (h) Δ1-AED/Δ6-T, (i) Δ1-AED/Δ15-AD, (j) Δ6-AED/Δ15-AD and (k) Δ6-T/Δ15-AD after the ingestion of 24 g or 42 g of EtOH.


Drug Testing

and Analysis

is in agreement with the hypothesis that after EtOH ingestion theT metabolism gets suppressed causing a decrease of androster-one and etiocholanolone excretion as previously reported. Boththe increase of testosterone and the decrease of androsterone andetiocholanolone is probably connectedwith a competitive inhibitionof some enzymes like 17ß-hydoxysteroid-dehydrogenases.[42,58]

However, the increase observed for Δ6-T concentrations afterthe ethanol ingestion is similar to the behaviour observed for Tconcentrations. This fact supports the hypothesis that Δ6-T is adirect product of T, whereas Δ1-AED and Δ6-AED most likely aremetabolites of androstendione.[14]

Since similar behaviour was observed for Δ1-AED, Δ6-AED andΔ15-AD, the use of ratios involving these steroidsminimized the var-iations (below 40%) after the EtOH ingestion (Figures 2g, 2i, and 2j).Due to its increase, the ratios involving Δ6-T exhibited higher varia-tions (Figures 2g and 2k).

Figure 3. The variations observed for the ratios (a) Andros/Etio, b) 5α-Adiol/5β-A(d) Δ1-AED, (e) Δ6-AED, (f) Δ6-T, (g) Δ15-AD; and the selected ratios between(k) Δ6-AED/Δ15-AD and (l) Δ6-T/Δ15-AD after the administration of a 5α-reduc


Thus, the use of the ratios Δ1-AED/Δ6-AED, Δ1-AED/Δ15-AD andΔ6-AED/Δ15-AD exhibited lower variations than those observedfor the T/E ratio after EtOH ingestion.

5α-reductase inhibitors

The effect of the use of 5α-reductase inhibitors was evaluated bymonitoring the variation of the concentrations of the cysteinyl com-pounds and the variation of the selected ratios.

The ratio between Andros and Etio (Figure 3a) and between5α-Adiol and 5β-Adiol (Figure 3b) was used for the evaluation ofthe 5α-reductase inhibitory effect. As expected, large decreases(around 7 and 5 times, respectively) were observed for theAndros/Etio and 5α-Adiol/5β-Adiol values after the administrationof dutasteride (Figures 3a and 3b). These results are in agreementwith the inhibition of the biosynthesis of the 5α-metabolites

diol and (c) T/E and the concentrations (ng/mL) of the cysteinyl compoundsthese compounds (h) Δ1-AED/Δ6-AED, (i) Δ1-AED/Δ6-T, (j) Δ1-AED/Δ15-AD,tase inhibitor.


117

A. Fabregat et al.

Drug Testing

and Analysis

118

(Andros and 5α-Adiol) by the 5α-reductase inhibitor and the con-comitant increase in Etio and 5β-Adiol productions, as previouslyreported.[51]

Regarding, the cysteinyl metabolites, as it can be seen inFigures 3d–3g, after the administration of dutasteride, an increasewas observed for the concentrations of Δ6-AED, Δ6-T and Δ15-AD(between 4-fold and 7-fold), compared to the increase of less than1 fold observed for Δ1-AED. Thus, the increase observed for thecysteinyl steroids is in agreement with the hypothesis that the inhi-bition of one of themajor T metabolic pathways leads to the activa-tion of minor metabolic pathways which increase the biosynthesisof Δ1-AED, Δ6-AED, Δ6-T and Δ15-AD.Regarding ratios, the excretion of larger amounts of Δ6-AED and

Δ6-T resulted in a 2-fold decrease in the Δ1-AED/Δ6-AED ratio and a3-fold decrease in the Δ1-AED/Δ6-T ratio (Figures 3g and 3i). SinceΔ15-AD rose to a lesser extent, all the ratios having Δ15-AD as thedenominator showed increments ranging from 2- to 7-fold afterthe drug intake (Figures 3j, 3k, and 3l).

Conclusions

The influence of the common factors that can alter the steroid pro-file has been studied for four steroids excreted in urine conjugatedwith cysteine; Δ1-AED, Δ6-AED, Δ6-T and Δ15-AD.These cysteinyl compounds were shown to be relatively stables

to those factors related to the sample preservation such asmicrobialdegradation and freeze/thaw cycles. The results observed for thesecompounds were similar than those observed for the T/E ratio.Regarding the endogenous factors influencing the excretion of

these compounds, similar effects to those observed when studyingthe T/E ratio were observed. Moderate infradian and circadianrhythms (with variations less than 30%)were found for the cysteinylsteroids. Gender variations on these metabolites were also re-vealed. In particular, higher concentrations of Δ6-AED and Δ6-Twere observed in female urine samples. In addition to that, alter-ations on the excretion of the cysteinyl compounds were observedduring the earliest stages of pregnancy. As expected, the variationsobserved in the excretion of T in Asian populations due to theUGT2B17 polymorphism do not have effects on the excretion ofthe steroids conjugated with cysteine. Thus, the ethnic variationswhich are one of the major drawbacks when using a general T/Ecut-off value can be circumvented by using ratios betweencysteinyl compounds.The use of substances which are known to influence the steroid

profile such as EtOH or 5α-reductase inhibitors led to variations sim-ilar to those observed for the T/E ratio. After the EtOH ingestion, allcysteinyl compounds exhibited a decrease in their concentrationsin the first hours after the intake. Only Δ6-T increased after EtOH ad-ministration for several hours. The use of 5α-reductase inhibitorsproduced an increase in the excretion of these compounds in urineprobably by the activation of minor metabolic pathways.Altogether, this study showed the potential of these compounds

to be implemented in the steroid profile. The inclusion of thecysteinyl compounds would potentially complement the currentsteroid profile allowing for the circumvention of some of itslimitations.

Acknowledgements

Financial support of WADA (11A22OP) is acknowledged. The au-thors would like to thank Magí Farré and Rafael de la Torre for pro-viding the samples of alcohol consumption.


References

[1] WADA. Anti-doping testing figures report. Available at: https://www.wada-main-prod.s3.amazonaws.com/resources/files/WADA-2013-Anti-Doping-Testing-Figures-LABORATORY-REPORT.pdf. [9 March 2015].

[2] M. Donike, K.-R. Bärwald, K. Klostermann, W. Schänzer, J. Zimmermann.Nachweis von exogem testosterone. in Sport: Leistung undGesundheit/Kongreßbd, (Eds: H. Heck, H. Hollman, H. Liesen, R. Rost),Deutscher Artze-Verlag: Köln, Germany, 1982, pp. 293.

[3] WADA. Technical document for Endogenous Anabolic Androgenic Ste-roids. Available at: https://www.wada-main-prod.s3.amazonaws.com/resources/files/wada-td2014eaas-v2-endogenous-anabolic-androgenic-steroids-measurement-and-reporting-en_0.pdf. [9 March 2015].

[4] M. Becchi, R. Aguilera, Y. Farizon, M. M. Flament, H. Casabianca,P. James. Gas chromatography/combustion/isotope-ratio massspectrometry analysis of urinary steroids to detect misuse oftestosterone in sport. Rapid Commun. Mass Spectrom. 1994, 8, 304.

[5] R. Aguilera, M. Becchi, H. Casabianca, C. K. Hatton, D. H. Catlin,B. Starcevic, H. G. Pope Jr. Improved method of detection oftestosterone abuse by gas chromatography/combustion/isotope ratiomass spectrometry analysis of urinary steroids. J. Mass Spectrom.1996, 31, 169.

[6] R. Aguilera, M. Becchi, C. Grenot, H. Casabianca, C. K. Hatton. Detectionof testosterone misuse: Comparison of two chromatographicsample preparation methods for gas chromatographic-combustion/isotope ratio mass spectrometric analysis. J. Chromatogr. B. 1996,687, 43.

[7] P. E. Sottas, M. Saugy, C. Saudan. Endogenous steroid profiling in theathlete biological passport. Endocrin. Metab. Clin. 2010, 39, 59.

[8] U. Mareck, H. Geyer, G. Opfermann, M. Thevis, W. Schanzer. Factorsinfluencing the steroid profile in doping control analysis. J. MassSpectrom. 2008, 43, 877.

[9] D. H. van de Kerkhof, D. de Boer, J. H. Thijssen, R. A. Maes. Evaluation oftestosterone/epitestosterone ratio influential factors as determined indoping analysis. J. Anal. Toxicol. 2000, 24, 102.

[10] P. Van Renterghem, P. Van Eenoo, W. Van Thuyne, H. Geyer,W. Schanzer, F. T. Delbeke. Validation of an extended method for thedetection of the misuse of endogenous steroids in sports, includingnew hydroxylated metabolites. J. Chromatogr. B Analyt. Technol.Biomed. Life Sci. 2008, 876, 225.

[11] P. Van Renterghem, P. Van Eenoo, F. T. Delbeke. Population basedevaluation of a multi-parametric steroid profiling on administeredendogenous steroids in single low dose. Steroids. 2010, 75, 1047.

[12] P. Van Renterghem, P. E. Sottas, M. Saugy, P. Van Eenoo. Statisticaldiscrimination of steroid profiles in doping control with supportvector machines. Anal. Chim. Acta. 2013, 768, 41.

[13] O. J. Pozo, J. Marcos, R. Ventura, A. Fabregat, J. Segura. Testosteronemetabolism revisited: Discovery of new metabolites. Anal. Bioanal.Chem. 2010, 398, 1759.

[14] A. Fabregat, A. Kotronoulas, J. Marcos, J. Joglar, I. Alfonso, J. Segura,R. Ventura, O. J. Pozo. Detection, synthesis and characterization ofmetabolites of steroid hormones conjugated with cysteine. Steroids.2013, 78, 327.

[15] O. J. Pozo, C. Gomez, J. Marcos, J. Segura, R. Ventura. Detection andcharacterization of urinary metabolites of boldione by LC-MS/MS.Part II: Conjugates with cysteine and N-acetylcysteine. Drug Test.Anal. 2012, 4, 786.

[16] A. Fabregat, O. J. Pozo, J. Marcos, J. Segura, R. Ventura. Quantification oftestosterone and metabolites released after alkaline treatment inhuman urine. Drug Test. Anal. 2010, 2, 630.

[17] A. Fabregat, O. J. Pozo, J. Marcos, J. Segura, R. Ventura. Alternativemarkers for the long-term detection of oral testosterone misuse.Steroids. 2011, 76, 1367.

[18] A. Fabregat, O. J. Pozo, P. Van Renterghem, P. Van Eenoo, J. Marcos,J. Segura, R. Ventura. Detection of dihydrotestosterone gel, oraldehydroepiandrosterone, and testosterone gel misuse through thequantification of testosterone metabolites released after alkalinetreatment. Drug Test. Anal. 2011, 3, 828.

[19] M. Tsivou, D. G. Georgakopoulos, H. A. Dimopoulou, M. A. Koupparis,J. Atta-Politou, C. G. Georgakopoulos. Stabilization of human urinedoping control samples: A current opinion. Anal. Bioanal. Chem.2011, 401, 553.

[20] M. E. Manson, L. Nocke-Fink, J. Å. Gustafsson, C. H. I. Shackleton. Artifactformation during incubation of neutral steroids with commercialβ-glucuronidase preparations. Clin. Chim. Acta. 1972, 38, 45.


https://www.wada-main-prod.s3.amazonaws.com/resources/files/wada-td2014eaas-v2-endogenous-anabolic-androgenic-steroids-measurement-and-reporting-en_0.pdf




Drug Testing

and Analysis

[21] G. Messeri, G. Cugnetto, G. Moneti, M. Serio. Helix pomatia inducedconversion of some 3 beta-hydroxysteroids. J. Steroid Biochem. 1984,20, 793.

[22] J. M. Kjeld, C. M. Puah, G. F. Joplin. Rise of unconjugated sex hormonesin human urine on storage. Clin. Chim. Acta. 1977, 80, 285.

[23] M. Tsivou, D. Livadara, D. G. Georgakopoulos, M. A. Koupparis,J. Atta-Politou, C. G. Georgakopoulos. Stabilization of human urinedoping control samples: II. microbial degradation of steroids. Anal.Biochem. 2009, 388, 146.

[24] C. Ayotte, D. Goudreault, A. Charlebois. Testing for natural andsynthetic anabolic agents in human urine. J. Chromatogr. B. 1996,687, 3.

[25] WADA. International Standards of Laboratories. Available at: https://www.wada-main-prod.s3.amazonaws.com/resources/files/WADA_Int_Standard_Laboratories_2012_EN.pdf. [9 March 2015].

[26] C. H. Shackleton. Profiling steroid hormones and urinary steroids.J. Chromatogr. 1986, 379, 91.

[27] H. Geyer, S. Rauth, B. Sample. Excretion of ephedrine and endogenoussteroids under conditions of controlled water intake and of waterdiuresis. in Recent Advances in Doping Analysis, (Eds: M. Donike, H.Geyer, U. Gotzmann, U. Mareck), Sports und Buch Straush: Köln,Germany, 1993, pp. 163.

[28] D. H. Catlin, C. K. Hatton, S. H. Starcevic. Issues in detecting abuse ofxenobiotic anabolic steroids and testosterone by analysis of athletes’urine. Clin. Chem. 1997, 43, 1280.

[29] J. Beanziger, L.D. Bowers. Variability of T/E ratios in athletes. in RecentAdvances in Doping Analysis, (Eds: M. Donike, H. Geyer, U. Gotzmann,U. Mareck, S. Rauth), Sport und Buch Strausse Edition Sport: Köln,Germany, 1993, pp. 41.

[30] H. Gall, H. J. Glowania, M. Fischer. Circadian rhythm of testosteronelevel in plasma. I. Physiologic 24-hour oscillations of the testosteronelevel in plasma. Andrologia. 1979, 11, 287.

[31] U. Mareck, H. Geyer, M. Donike. Stability of steroids profiles(3): Thecircadian rythm of urinary ratios and excretion rates of endogenoussteroids in male. in Recent Advanced in Doping Analysis, (Eds: M.Donike, H. Geyer, U. Gotzmann, U. Mareck), Sport und Buch StrausseEdition Sport: Köln, Germany, 1995, pp. 177.

[32] U. Mareck, H. Geyer, M. Donike. Stability of steroid profiles (4): Thecircadian rhythm of urinary ratios and excretion rates of endogenoussteroids in female and its menstrual dependency. in Recent Advancesin Doping Analysis, (Eds: M. Donike, H. Geyer, U. Gotzmann, U. Mareck),Sport und Buch Strausse Edition Sport: Köln, Germany, 1995, pp. 135.

[33] B. K. Yap, R. Kazlauskas, K. Elghazi, G. A. Johnston, R. P. Weatherby.Profiling of urinary testosterone and luteinizing hormone in exercise-stressed male athletes, using gas chromatography–mass spectrometryand enzyme immunoassay techniques. J. Chromatogr. B. 1996, 687, 117.

[34] M. Hill, A. Parizek, D. Cibula, R. Kancheva, J. E. Jirasek, M. Jirkovska,M. Velikova, J. Kubatova, M. Klimkova, A. Paskova, Z. Zizka,L. Kancheva, H. Kazihnitkova, L. Zamrazilova, L. Starka. Steroidmetabolome in fetal and maternal body fluids in human latepregnancy. J. Steroid Biochem. 2010, 122, 114.

[35] A. Fabregat, J. Marcos, L. Garrostas, J. Segura, O. J. Pozo, R. Ventura.Evaluation of urinary excretion of androgens conjugated to cysteinein human pregnancy by mass spectrometry. J. Steroid Biochem. 2014,139, 192.

[36] M. E. van Houten, L. J. Gooren. Differences in reproductiveendocrinology between Asian men and Caucasian men--a literaturereview. Asian J. Androl. 2000, 2, 13.

[37] J. Jakobsson, L. Ekstrom, N. Inotsume, M. Garle, M. Lorentzon,C. Ohlsson, H. K. Roh, K. Carlstrom, A. Rane. Large differences intestosterone excretion in Korean and Swedish men are stronglyassociated with a UDP-glucuronosyl transferase 2B17 polymorphism.J. Clin. Endocrinol. Metab. 2006, 91, 687.

[38] P. Anielski, J. Simmchen, L. Wassill, D. Ganghofner, D. Thieme.Epidemiological investigation of the UGT2B17 polymorphism indoping control urine samples and its correlation to T/E ratios. DrugTest. Anal. 2011, 3, 645.

[39] C. Ayotte. Detecting the administration of endogenous anabolicandrogenic steroids. Handb. Exp. Pharmacol. 2010, 195, 77.


[40] J. J. Schulze, J. Lundmark, M. Garle, I. Skilving, L. Ekstrom, A. Rane. Dopingtest results dependent on genotype of uridine diphospho-glucuronosyltransferase 2B17, the major enzyme for testosterone glucuronidation.J. Clin. Endocrinol. Metab. 2008, 93, 2500.

[41] A. A.-B. Badawy. The metabolism of alcohol. Clin. Endocrinol. Metab.1978, 7, 247.

[42] T. Karila, V. Kosunen, A. Leinonen, R. Tahtela, T. Seppala. High doses ofalcohol increase urinary testosterone-to-epitestosterone ratio infemales. J. Chromatogr. B. 1996, 687, 109.

[43] O. Falk, E. Palonek, I. Bjorkhem. Effect of ethanol on the ratio betweentestosterone and epitestosterone in urine. Clin. Chem. 1988, 34, 1462.

[44] S. H. Andersson, T. Cronholm, J. Sjovall. Effects of ethanol on the levelsof unconjugated and conjugated androgens and estrogens in plasmaof men. J. Steroid Biochem. 1986, 24, 1193.

[45] M. J. Valimaki, M. Harkonen, C. J. Eriksson, R. H. Ylikahri. Sex hormonesand adrenocortical steroids in men acutely intoxicated with ethanol.Alcohol. 1984, 1, 89.

[46] W. Weinmann, P. Schaefer, A. Thierauf, A. Schreiber, F. M. Wurst.Confirmatory analysis of ethylglucuronide in urine by liquid-chromatography/electrospray ionization/tandem mass spectrometryaccording to forensic guidelines. J. Am. Soc. Mass Spectrom. 2004,15, 188.

[47] S. Dresen, W. Weinmann, F. M. Wurst. Forensic confirmatory analysis ofethyl sulfate--a new marker for alcohol consumption--by liquid-chromatography/electrospray ionization/tandem mass spectrometry.J. Am. Soc. Mass Spectrom. 2004, 15, 1644.

[48] D. Thieme, J. Grosse, L. Keller, M. Graw. Urinary concentrations of ethylglucuronide and ethyl sulfate as thresholds to determine potentialethanol-induced alteration of steroid profiles. Drug Test. Anal. 2011,3, 851.

[49] R. S. Rittmaster. 5alpha-reductase inhibitors in benign prostatichyperplasia and prostate cancer risk reduction. Best Pract. Res. Cl. En.2008, 22, 389.

[50] E. A. Olsen, M. Hordinsky, D. Whiting, D. Stough, S. Hobbs, M. L. Ellis,T. Wilson, R. S. Rittmaster. The importance of dual 5α-reductaseinhibition in the treatment of male pattern hair loss: Results of arandomized placebo-controlled study of dutasteride versusfinasteride. J. Am. Acad. Dermatol. 2006, 55, 1014.

[51] M. Thevis, H. Geyer, U. Mareck, U. Flenker, W. Schanzer. Doping-controlanalysis of the 5alpha-reductase inhibitor finasteride: Determination ofits influence on urinary steroid profiles and detection of its majorurinary metabolite. Ther. Drug Monit. 2007, 29, 236.

[52] H. Zimmer, G. Schmitt, R. Aderjan. Preliminary immunochemical test forthe determination of ethyl glucuronide in serum and urine:Comparison of screening method results with gas chromatography–mass spectrometry. J. Anal. Toxicol. 2002, 26, 11.

[53] WADA. Decision limits for the confirmatory quantification of thresholdsubstances. Available at: https://www.wada-main-prod.s3.amazonaws.com/resources/files/WADA-TD2014DL-v1-Decision-Limits-for-the-Quantification-of-Threshold-Substances-EN.pdf. [9 March 2015].

[54] M. Mazzarino, M. G. Abate, R. Alocci, F. Rossi, R. Stinchelli, F. Molaioni,X. de la Torre, F. Botrè. Urine stability and steroid profile: Towards ascreening index of urine sample degradation for anti-dopingpurpose. Anal. Chim. Acta. 2011, 683, 221.

[55] C. Jimenez, R. de la Torre, J. Segura, R. Ventura. Stability studies oftestosterone and epitestosterone glucuronides in urine. RapidCommun. Mass Spectrom. 2006, 20, 858.

[56] R. de la Torre, X. de la Torre, J. Segura, M. Smeyers, R. Ventura, J. Torres,C. Alia, T. Baró. Urine contamination bymicroorganisms and alterationsin the endogenous steroid profile. in Recent Advances in DopingAnalysis, (Eds: M. Donike, H. Geyer, U. Gotzmann, U. Mareck), Sportsund Buch Straush: Köln, Germany, 1997, pp. 223.

[57] U. Mareck-Engelke, H. Geyer, M. Donike, Stability of steroid profiles. inRecent Advances in Doping Analysis, (Eds: M. Donike, H. Geyer, U.Gotzmann, U. Mareck, S. Rauth), Sport und Buch Strausse EditionSport: Köln, Germany, 1993, pp. 87.

[58] J. Grosse, P. Anielski, H. Sachs, D. Thieme. Ethylglucuronide as apotential marker for alcohol-induced elevation of urinarytestosterone/epitestosterone ratios. Drug Test. Anal. 2009, 1, 526.


119

https://www.wada-main-prod.s3.amazonaws.com/resources/files/WADA-TD2014DL-v1-Decision-Limits-for-the-Quantification-of-Threshold-Substances-EN.pdf



Factors affecting urinary excretion of testosterone

Documents

Transcript of Factors affecting urinary excretion of testosterone