Anal Bioanal Chem (2011) 401:553–561 DOI 10.1007/s00216-011-4887-5
ORIGINAL PAPER
Stabilization of human urine doping control samples: a current opinion Maria Tsivou & Dimitrios G. Georgakopoulos & Helen A. Dimopoulou & Michael Α. Koupparis & Julia Atta-Politou & Costas G. Georgakopoulos
Received: 5 January 2011 / Revised: 28 February 2011 / Accepted: 8 March 2011 / Published online: 16 April 2011 # Springer-Verlag 2011
Abstract Transportation of doping control urine samples from the collection sites to the World Anti-doping Agency (WADA) Accredited Laboratories is conducted under ambient temperatures. When sample delivery is not immediate, microbial contamination of urine, especially in summer, is a common phenomenon that may affect sample integrity and may result in misinterpretation of analytical data. Furthermore, the possibility of intentional contamination of sports samples during collection with proteolytic enzymes, masking the abuse of prohibited proteins such as erythropoietin (EPO) and peptide hormones, is a practice that has already been reported. Consequently, stabilization of urine samples with a suitable method in a way that protects samples’ integrity is important. Currently, no stabilization method is applied in the sample collection equipment system in order to prevent degradation of urine compounds. The present work is an overview of a study, Published in the special issue Anti-Doping Analysis with Guest Editor Mario Thevis. M. Tsivou : H. A. Dimopoulou : C. G. Georgakopoulos (*) Doping Control Laboratory of Athens, OAKA, Kifissias 37, 15123 Maroussi, Greece e-mail:
[email protected] D. G. Georgakopoulos Laboratory of General and Agricultural Microbiology, Faculty of Agricultural Biotechnology, Agricultural University of Athens, Iera Odos 75, 11855 Athens, Greece M. Tsivou : M. Α. Koupparis : J. Atta-Politou Laboratory of Analytical Chemistry, Chemistry Department, University of Athens, Panepistimioupolis, 15771 Athens, Greece
funded by WADA, on degradation and stabilization aspects of sports urine samples against the above threats of degradation. Extensive method development resulted in the creation of a mixture of chemical agents for the stabilization of urine. Evaluation of results demonstrated that the stabilization mixture could stabilize endogenous steroids, recombinant EPO, and human chorionic gonadotropin in almost the entire range of the experimental conditions tested. Keywords Urine stabilization . Doping-control analysis . Microorganisms . Endogenous steroids . Recombinant erythropoietin . Human chorionic gonadotropin
Introduction It is a widely accepted practice by the National Anti-doping Organizations (NADOs) and International Federations (IFs) to transport doping control urine samples from the collection sites to the World Anti-doping Agency (WADA) Accredited Laboratories without protection from the environmental conditions. Criticism against questionable WADA’s adverse analytical findings has relied on sample storage and transportation conditions [1, 2]. Several past studies have demonstrated the adverse effects of delays and improper storage conditions during transportation of urine samples [3–10]. Endogenous steroids can be degraded by microorganisms likely present in the human body or the environment, especially during the warm periods of the year [3–5, 7, 8, 11–13]. WADA has established criteria to check for signs of microbial degradation, such as the elevated amounts of 5α-androstane-3,17-dione (5α-dione), 5β-androstane-3,17-dione (5β-dione) in the free form and the presence of free testosterone or epitestosterone at a
554
M. Tsivou et al.
percentage exceeding 5% of the respective glucuroconjugates [14]. Recently, microbial contaminants have been characterized in doping-control samples with polymerase chain reaction techniques [15]. Besides, reports have been published regarding the on-purpose degradation of proteins in urine, such as human chorionic gonadotropin (hCG), recombinant EPO (rEPO) and insulins, by athletes during the sample collection procedure [16–19]. Urine adulteration by proteases was recently included in WADA’s prohibited list under the category of chemical and physical manipulation [20]. Moreover, the occurrence of endogenously produced proteases that are renally excreted from the human body has been investigated by Thevis et al. [16]. An effective stabilization method that prevents or minimizes urinary steroid and hormones degradation is suggested as a way to overcome this challenge [8, 16, 21, 22]. A variety of stabilization agents have been tested in the field of anti-doping science to inhibit the formation of steroid degradation products or increase the stability of EPO in urine [4, 6, 10, 21, 22]. More specifically, bacteriostatic agents such as sodium azide (NaN3) and mercuric chloride (HgCl2) have been added to urine samples to inhibit the formation of 5α-dione and 5β-dione as well as the deconjugation of urinary steroids [4, 6, 23]. Urinary EPO was reported to be stable even at room temperature in the presence of NaN3 [24]. Concerning the inactivation of proteolytic enzymes in urine, a protease detection kit was used to test the effectiveness of different serine protease inhibitors [25]. However none of the tested inhibitors were able to fully inhibit the catalytic activity of subtilisin A. EPO digestion by trypsin was prevented using a combination of protease inhibitors such as “Complete” (Roche Diagnostics GmbH, Manheim, Germany) and pepstatin [22]. No results were reported regarding the chemical stabilization of hCG in urine specimens against proteolytic and thermal degradation for the purposes of Table 1 Current composition of the chemical stabilization mixture
Name
Substance
Concentration in urine
Sodium azide Antibiotic/antimycotic mixture
NaN3 Penicillin G sodium
10 mg/ml 2 ml/100 ml urine
Protease inhibitor cocktail With kind permission from ref. [37] a
(4-(2-Aminoethyl) benzenesulfonyl fluoride b
[N-[N-[L-3-trans-Carboxirane-2carboxyl]-L-leucyl]-agmatine]
c d
Ethylenediaminetetraacetic acid, Phenylmethylsulfonyl fluoride
doping control analysis. The only information available was related to sample handling issues of gonadotropins in doping [26]. Up to the present time, no preservative is added to sport urine samples [27, 28] because, on one hand, it is reasoned that the introduction of chemical substances into athletes’ samples after the collection procedure may open doors to litigation [6, 8, 27]. On the other hand, ethylenediaminetetraacetic acid (EDTA) is included in whole blood or plasma collection tubes as an anticoagulant [29–31] whereas an inert serum separator gel and a clotting activation factor is used in serum collection tubes [29, 32, 33] without the raise of any relevant criticism. This project was undertaken within the context of a WADA research grant investigating the efficiency of a chemical mixture in a series of incubation experiments to stabilize the endogenous steroids profile and prevent digestion of proteins like rEPO and hCG in case of microbial or enzymatic action in urine samples. Physical stabilization methods were found to be ineffective either for their practical application or for the inhibition of both microbial and proteolytic action [34]. The extensive method development performed revealed that there is no uniform chemical agent that is able to inactivate a wide range of microorganisms and proteolytic enzymes at the same time. This fact led to the creation of a mixture of antibiotic, antimycotic substances, and protease inhibitors, presented in Table 1. Apart from the stabilizing ability of the mixture’s constituents, other selection criteria were the low cost, the toxicity and the absence of matrix interferences in the doping control analysis. Incubation experiments were conducted for the stabilization of endogenous steroids and the proteins rEPO and hCG. The results reported in [35–37] demonstrated that in the presence of the chemical stabilization mixture: (a) microbial growth was completely inhibited, (b) the hydrolysis of glucuronide
Specific serine protease inhibitor Specific pepsin inhibitor Specific trypsin inhibitor
Streptomycin sulfate Amphotericin B AEBSFa E−64b Bestatin.HCl Leupeptin.HCl Aprotinin EDTA.2Nac PMSFd Pepstatin A Soybean trypsin inhibitor
2 g cocktail/g protease
5 mM 4.6 mM 2 g trypsin inhibitor/g trypsin
Stabilization of human urine doping control samples
and sulfate steroid conjugates, formation of metabolic byproducts, and production of boldenone induced by certain microorganisms were prevented, (c) the electrophoretic signal of EPO was improved in stabilized urine even if proteolytic enzymes were added, in almost the entire range of the experimental conditions tested, and (d) the degradation of hCG induced by four of the proteases applied was prevented in stabilized aliquots at the end of the incubation period. These findings are promising for the future implementation of a specially designed sample collection container incorporating the chemical stabilization mixture. However, the minimization of analytical matrix interferences in the Accredited Laboratories screening procedures will have to be ensured before the implementation of the stabilization mixture in a wide scale. This constitutes the purpose of an on-going research project, funded by WADA.
Methods Experimental protocol A simplified flow chart describing the experimental protocol universally used in the project is depicted in Fig. 1. Pooled urine fortified with reference steroids (Steraloids, Newport, RI, USA) or rEPO (BRP European Pharmacopoeia Commission, Strasbourg, France) or intact hCG devoid of nicked forms and free subunits (hCG99/ 688, NIBSC, Hertfordshire, UK) was inoculated with microbial strains or spiked with proteolytic enzymes in the presence and absence of the stabilization mixture. All series of aliquots were incubated at 37 °C and in parallel stored at −20 °C to serve as reference. Seven prokaryotic (bacteria) and two eukaryotic (fungus and yeast) microorganisms, including species related with human microbial flora [15, 38, 39], urinary tract infections [15, 40], and indoor air pollution [41], were selected for the experimental protocol concerning endogenous steroids. The criteria for the selection of microorganisms are presented explicitly in [35]. Similarly the six proteases used in the experimental protocol referring to the stabilization of rEPO and hCG were selected according to their availability and their potential to be misused by athletes [16–19, 22, 25]. GC/ MS analyses in SIM and full scan modes were performed at t=0 and t=7 days for the steroid profile evaluation [35]. EPO concentrations were determined at different time points of the period investigated with an automated chemiluminescent assay (Immulite EPO assay, Siemens, Llanberis, Gwynedd, UK) (t=0, 1, 4, 7 days) [36] as well as the EPO test adopted by WADA based on isoelectric focusing (IEF), double-blotting and chemiluminescence detection [42] (t=0 and 4 days). For the monitoring of hCG degradation, after examination of several combina-
555
tions of immunoassay kits, and hCG variants, the EIAgen Total hCG kit (Adaltis Italia S.p.A, Bologne, Italy) was used because of its selectivity against intact hCG molecule and not to its hCGβ1 fragment [37]. Chemical stabilization mixture The mixture of chemical agents applied consists of antimicrobial substances in order to inactivate a wide range of microorganisms and of several protease inhibitors for the simultaneous inactivation of a wide range of proteases. The current composition of the chemical stabilization mixture (Table 1) encompasses NaN3 (Sigma-Aldrich Chemie, Steinheim, Germany) due to its bactericidal properties, a commercial antimicrobial liquid mixture (pen-strep-fungizone) with wide antibacterial and antimycotic spectra— against Gram-positive and Gram-negative bacteria, fungi, and yeasts (Gibco, Invitrogen, Grand Island, NY, USA) as well as several protease inhibitors. Specifically, the selected protease inhibitor cocktail comprises AEBSF, E−64, bestatin.HCl, leupeptin.HCl, aprotinin, and EDTA.2Na (SigmaAldrich Chemie, Steinheim, Germany). Even though it possesses a broad specificity for the inhibition of serine, cysteine, and aspartic proteases and metalloproteases, it was not found sufficient to impede proteolytic action in urine specimens. Therefore, PMSF (Sigma-Aldrich Chemie, Steinheim, Germany), an irreversible serine protease inhibitor, was also included. Moreover, the chemical mixture was fortified with pepstatin (Sigma-Aldrich Chemie, Steinheim, Germany), a highly specific inhibitor of acid proteases and proteases of microbial origin. Finally, the addition of a specific trypsin inhibitor from Glycine max (soybean) (Sigma-Aldrich Chemie, Steinheim, Germany) was considered necessary. Microbiological and statistical analysis of urine samples showing signs of microbial contamination Eleven urine samples, collected in July 2008, were submitted to microbiological analysis at the Department of Microbiology (Medical School of the University of Athens). The main selection criteria were signs of microbial degradation based on the elevated levels of 5α-dione following the results of the screening procedure for steroid profile analysis. Over a period of 2 years (May 2008–March 2010) routine urine samples showing formation of 5α-dione (with peak heights higher than 3×103) were recorded. Also,
1 β-subunit of hCG, according to the nomenclature proposed by the International Federation of Clinical Chemistry and Laboratory Medicine (IFCC) [43].
556
M. Tsivou et al.
Fig. 1 General flow chart of the experimental stabilization protocol
pooled urine from a young girl stored at -20 °C
urine aliquots
with addition of the chemical stabilization mixture
without addition of the chemical stabilization mixture negative controls
spiking with reference steroids, rEPO or hCG 99/688
spiking with reference steroids, rEPO or hCG 99/688
inoculation with microorganisms or spiking with proteases
without microorganisms or proteases
without microorganisms or proteases
inoculation with microorganisms or spiking with proteases
t = 0 : microbial growth, measurement of pH and sg steroid profile analysis, EPO analysis, hCG measurement
incubation at 37 °C and storage at -20 °C for 4 or 7 days
t = 4 or 7 days : microbial growth, measurement of pH and sg steroid profile analysis, EPO analysis, hCG measurement
12.0 t=0 without stabilizers t=7 without stabilizers
10.0
t=0 with stabilizers
10.0
Mean population of microorganisms (log cfu/ml +SD)
t=7 with stabilizers
(bar does not appear due to zero value)
8.5
7.8
8.0 7.5 7.6 7.1
7.1
8.0
7.3
6.9
6.8
6.7 6.2
5.9
6.3
5.6
5.3
5.3
6.0
6.1
5.2
4.5
4.8
4.9
4.4
4.0
2.0
0
0
0
0
0
0
0
0
0
0
0
0
s vu la A .f
an s C.
al b
ic
is ec al fa
pl N .s
im
m er id ep S.
E.
ex
s id i
lis ira bi m P.
on ne .p K
P.
ae r
um
ug
E.
in
co
os
ia e
a
li
0.0
Microorganisms
Fig. 2 Mean population of microorganisms (log10 cfu/ml+SD) in the beginning and at the end of the 7-day incubation period at 37 °C, in the presence and absence of the chemical stabilization mixture
Stabilization of human urine doping control samples
557
Fig.
3 Percentages of free androsterone (a), etiocholanolone (b), epitestosterone (c), and dehydroepiandrosterone (DHEA) (d) after a 7day incubation period at 37 °C with different microoganisms plotted on logarithmic scale. The numbers on the x-axis refer to the microroganisms listed in the panels. The deconjugation rates attributed to thermal hydrolysis of steroid conjugates are plotted first on the xaxis. Open triangles correspond to rates obtained in the untreated samples, whereas solid circles represent the rates observed with the addition of the chemical stabilization mixture. Reprinted from ref. [35], with permission from Elsevier
A Percentage of free androsterone
100,0 1. E. coli 2. P. aeruginosa 3. K. pneumoniae 4. P. mirabilis 5. S. epidermidis 6. N. simplex 7. E. faecalis 8. C. albicans 9. A. flavus
10,0
1,0 Thermal
1
2
3
0,1
4
5
6
7
8
9
information regarding the sampling sites and dates, gender, pH, and specific gravity (sg) values of the above-mentioned urine samples was gathered.
Microorganisms
B
Percentage of free etiocholanolone
100.0
Results 10.0
Analysis of degradation parameters from doping control urine samples 1.0 Thermal
1
2
3
0.1
4
5
6
7
8
9
Microorganisms
C Percentage of free epitestosterone
100.0
10.0
1.0 Thermal
1
2
3
0.1
4
5
6
7
8
9
Microorganisms
D
Percentage of free DHEA
100.0
Microbiological analysis of urine samples indicated the presence of Enterococcus spp. at levels higher than 1× 104 colony forming units (cfu)/ml in five out of eleven samples whereas two of the samples were contaminated with coagulase-negative staphylococci at levels higher than 1×105 cfu/ml. Three samples were found sterile. No yeast cells were detected in the same samples. From the total of 7,450 urine specimens received for analysis during the 2-year period investigated, approximately 3.7% were urines with bacterial activity. This percentage of contaminated samples is in accordance with those published by other authors [13, 21]. A part of 55% of the contaminated urines came from female athletes. The occurrence of degraded urine samples in Athens Doping Control Laboratory varied over the course of the year with a peak during the months of May, June, and July. Samples received for analysis were collected in Greece but also abroad. On some occasions, more than 6 days passed from the date of sample collection until the date of receipt at the laboratory. It has been demonstrated from previous studies [5, 7, 13, 15, 35] that alkalanization of urine is not a reliable indicator of microbial contamination. In the present study, only 24% of the samples showing formation of 5α-dione had elevated pH values (>8.0). Conversely, all samples with pH values above 8.0 were characterized by elevated levels of 5α-dione (peak heights higher than 3×103).
10.0
Stabilization of endogenous steroids
1.0 Thermal 1
2
3
4
5
Microorganisms
6
7
8
9
The effectiveness of the chemical mixture in preventing the degradation of endogenous steroids in urine was assessed after a 7-day incubation period at 37 °C inoculated with microbial species [35]. Evaluation of results showed that
558
M. Tsivou et al.
b)
a)
35
controls
Mean EPO concentration (mIU/ml)
Mean EPO concentration (mIU/ml)
70 60 50 40 30 20
trypsin
30 25 20 15 10 5
10
0
0 0
1
4
0
7
1
c)
50
Mean EPO concentration (mIU/ml)
Mean EPO concentration (mIU/ml)
bromelain 50 40 30 20 10
papain
45 40 35 30 25 20 15 10 5 0
0 0
1
4
0
7
1
days
4
7
days
e)
f) 16
subtilisin A
60
Mean EPO concentration (mIU/ml)
Mean EPO concentration (mIU/ml)
7
d) 60
14 12 10 8 6 4
0
50 40 30 20
0 0
1
days
4
7
0
g) 18
pepsin
10
2
Mean EPO concentration (mIU/ml)
4
days
days
-chymotrypsin
16 14 12 10 8 6 4 2 0 0
1
4
days
7
1
days
4
7
Stabilization of human urine doping control samples
559
Fig. 4
Mean EPO concentration (n=3) a of control samples and in the presence of 100 μg/ml of proteases: b trypsin, c bromelain, d papain, e subtilisin A, f pepsin, and g α-chymotrypsin during a 7-day incubation period at 37 °C and parallel storage at −20 °C. Empty triangles, incubation at 37 °C without the addition of the chemical stabilization mixture; empty circles: storage at −20 °C without the addition of the chemical stabilization mixture; filled triangles, incubation at 37 °C in the presence of the chemical stabilization mixture; filled circles, storage at −20 °C in the presence of the chemical stabilization mixture
the introduction of the chemical stabilization mixture in urine aliquots inhibited the microorganisms’ growth and prevented steroid degradation at 37 °C. Specifically, cell growth was completely impeded in urine samples treated with the stabilization mixture even after a 7-day incubation period at 37 °C (Fig. 2). In the untreated aliquots, the microorganisms did not only survive but also the number of colonies increased over time under the favorable temperature of 37 °C except for Proteus mirabilis, Klebsiella pneumoniae, and Aspergillus flavus. Four of the nine microorganisms induced alterations in the steroid profile of the unstabilized samples: Escherichia coli, Nocardioides simplex, Candida albicans, and A. flavus. The rest of the microorganisms produced minor changes in the steroid profile. Three microorganisms induced bacterial deconjugation of steroids after the 7-day incubation period at 37 °C (outlying values in Fig. 3 panels). In urine aliquots contaminated with E. coli, N. simplex, A. flavus, and C. albicans, synthesis of 5α-dione and 5β-dione was observed at the end of the incubation period at 37 °C. However, the addition of the chemical mixture to urine aliquots stopped microbial degradation, and no formation of 5α-dione and
5β-dione was detected at the end of the incubation period at 37 °C. The pH and sg values of the urine samples were not significantly affected by the addition of the chemical mixture [35]. Stabilization of rEPO The possibility of preventing EPO digestion in urine samples by six common proteases was investigated using the chemical stabilization mixture [36]. After EPO analysis with Immulite EPO assay, it was found that the presence of six proteases in the urine aliquots untreated with the stabilization mixture resulted in reduced EPO levels over time, especially following incubation at 37 °C (Fig. 4 panels). In the presence of the chemical stabilization mixture, EPO levels were significantly higher especially after 4 days of incubation at 37 °C in aliquots containing trypsin, bromelain, papain and α-chymotrypsin compared with the untreated aliquots (Fig. 4b, c, d, f, respectively). In the presence of subA, EPO content did not differ significantly between aliquots treated with the chemical stabilization mixture and the untreated aliquots after the 4day incubation period at 37 °C. The time of sample exposure to proteases seems a critical parameter for the inhibition of EPO degradation using the chemical mixture. When the incubation time was 7 days instead of 4 days, the concentration of EPO was significantly higher in the presence of the chemical stabilization mixture only for bromelain and papain compared with the untreated aliquots. When the EPO test, based on IEF, was applied for qualitative EPO analysis, visible bands in the EPO A
180
t=0 without stabilizers t=4 days without stabilizers t=0 with stabilizers t=4 days with stabilizers
171
160
B
140
132
128
hCG (mIU/ml)
A
120
B
A A B
100 85 B
79
80 60
100
B 59 B
91 78
B
81
76
86
85
75 69
67
A
64 55
A
40
97
92
90
B
42
34 27
20
16 0
5
1
0
subA
trypsin
-chymotrypsin
Fig. 5 Changes in hCG concentration (mIU/ml) due to protease addition with and without the chemical stabilization mixture following a 4-day incubation period at 37 °C. A, groups significantly different at
papain
bromelain
pepsin
controls
t=0, p
