| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Doping in Sport Symposium Proceedings |
Athletic Drug Testing and Toxicology Laboratory, Department of Pathology and Laboratory Medicine, Indiana University Medical Center, 635 Barnhill Dr., Indianapolis, IN 46202-5120. Fax 317-274-7641; e-mail lbowers{at}iupui.edu
Abstract
The use and abuse of performance-enhancing substances has been an issue in sports since the ancient Greeks. The availability of numerous synthetic steroids and recombinant peptide hormones has made testing an analytical challenge. Recent advances in mass spectrometry have provided an opportunity to decrease detection limits. The Atlanta Olympic Games in 1996 marked the first time every specimen was screened by gas chromatography (GC) coupled to high-resolution mass spectrometry (MS). A further improvement may be seen with GC/MS/MS and quadrupole ion traps. Electrospray HPLC/MS has also been applied to the detection and confirmation of peptide hormones in urine. The ability to detect subtle differences in oligosaccharide structure may provide a way to detect abuse of recombinant glycoproteins. Simply decreasing detection limits is not enough; new technology also allows development of a foundation on which to base interpretation. Application of HPLC/MS/MS has allowed direct measurement of steroid conjugates in urine. The relative importance of sulfate, glucuronide, and other conjugates and metabolites of testosterone and epitestosterone can now be assessed. In the international sports arena, the impact of genetic metabolic disposition must also be considered if we are to provide an equitable system. Further research will establish more-refined criteria for the detection threshold of abused substances.
Key Words: indexing terms: sports medicine • abused drugs • drug assays • GC-MS • tandem MS • steroids • peptide hormones • urine • genetic variants
The use of performance-enhancing drugs in sport has been controversial since the time of the ancient Greeks. In the early 1960s, the Council of Europe banned the use of endogenous or exogenous agents to achieve an artificial and unfair increase in performance in competition. In general, the rationale for banning certain pharmaceutical agents is the desire for a "level playing field" and concern for the health of the athlete (1). The attitude of some athletes, "The game... is to always be one up on the testers" (2), makes continued improvement of testing imperative if a level playing field is to be assured. The International Olympic Committee (IOC)2 has an extensive list of banned substances fitting into the general categories of stimulants, narcotic analgesics, anabolic agents, peptide hormones, and urine-manipulating agents.
The first testing for banned substances was carried out at the 1972 Olympic Games in Munich, where gas chromatography with nitrogen-selective detection was used to test >2000 urine specimens for stimulants. Although GC/MS was used to confirm the presence of anabolic steroids at the 1976 Montreal Olympic Games, a major advance in analytical methodology was implemented at the 1983 Pan American Games, where benchtop quadrupole GC/MS instruments were used to screen all urine specimens. Another potentially significant advance in analytical technology took place the 1996 Olympic Games in Atlanta, where GS/sector MS systems were used at moderate mass resolution to screen all of the urine specimens for anabolic agents.
With the ever-increasing number of dietary supplements used by athletes, as well as the pharmacological sophistication of their trainers and physicians, detection of performance-enhancing compounds banned by the IOC is increasingly difficult. The combination of "no-notice" testing of athletes (while out of competition) and at-competition testing with appropriate analytical approaches can minimize the opportunity to achieve benefit from banned substances. Here, I review some of the latest advances in analytical methods under development to test for drugs that affect athletic performance.
Materials and Methods
All materials were used as obtained without further purification.
Sulfate and glucuronide conjugates of unlabeled (d0)
testosterone were purchased from Steraloids (Wilton, NH). Unlabeled
(d0) epitestosterone conjugates and all four
16,16,17-1
H3-labeled (d3)
conjugates (used as internal standards) were synthesized in our
laboratory. Ammonium acetate (purity 99.999%) and acetic acid (doubly
distilled) were from Aldrich (Milwaukee, WI). Human growth hormone
(hGH) preparations [Humatrope (Lilly, Indianapolis, IN) and Protopin
(Genentech, South San Francisco, CA)] were obtained from the Indiana
University Hospital Pharmacy. Insulin-like growth factor-1 (IGF-1) was
obtained from Bachem Bioscience (King of Prussia, PA). High-purity (18
M
) water was obtained by passing house-distilled deionized water
through a Barnstead Nanopure II system (Syborn/Barnstead, Boston, MA).
HPLC-grade methanol was obtained from Curtin Matheson (Houston, TX).
All MS/MS work was carried out with a PE–Sciex (Perkin-Elmer–Sciex, Norwalk, CT) API IIIPLUS triple-quadrupole MS equipped with IonsprayTM, a pneumatically assisted electrospray (ES) interface. To introduce the HPLC effluent, this laboratory uses a Beckman (Brea, CA) Model 126 programmable solvent module used in conjunction with a Rheodyne (Cotati, CA) 8125 injection valve housed in a DuPont (Wilmington, DE) forced-air oven. Quantitative data acquisition was done with the Routine Acquisition and Display program and the areas measured with the MACQUAN program (both PE-Sciex). Details of the experiments can be found in the relevant references.
Results and Discussion
anabolic steroids
Recent statistics from the IOC-accredited laboratories indicate
that anabolic agents remain the primary performance-enhancing
substances detected in athletes. A review of anabolic steroid
metabolism has been published recently (3). The analytical
challenge in urine steroid analysis is threefold: the low
concentrations of anabolic steroid metabolites (µg/L range); the
large number of steroids and steroid metabolites sought; and the
complexity of the urine matrix, which contains steroids of similar
structure at concentrations as much as 1000-fold higher than the
compounds of interest. A flow diagram of a commonly used approach for
the analysis of anabolic agents is shown in Fig. 1
; note that only the glucuronide conjugates of the steroids are
routinely analyzed. The requirement for low limits of detection (LOD;
100 pg on-column) has mandated the use of selected-ion monitoring (SIM)
GC/MS. A typical SIM method for anabolic agents incorporates 10–12
time-programed acquisition groups of 10–20 ions each. Sequential
isolation of this many ions, along with the incumbent instrument
overhead time, is hindered by the sampling and signal-to-noise
limitations of the mass filter.
|
In principle, the quadrupole ion trap (QIT) has an advantage over the quadrupole mass filter when large numbers of ions must be monitored. The QIT can be operated in three modes: selected ion detection; selected ion storage; and selected ion ejection. In selected ion storage, an ion of a single mass-to-charge ratio (m/z) is retained in the QIT at a time, much like SIM operation of a quadrupole mass filter. In the selected ion ejection mode, several ions of different m/z values can be stored simultaneously. The advantage of filling the QIT with only the ions of interest is improved limits of detection. Unfortunately, selected ion ejection techniques appear to be of limited benefit when the background is determined by the chemical composition of the biological matrix. In this case, the relative signal arising from the compound of interest can be increased, but the signal-to-noise ratio is not improved, given a similar increase in the background signal from compounds of similar mass [4].
LOD can be improved through more-selective detection techniques, such
as GC/high-resolution mass spectrometry (HRMS). Mass resolution (R) is
defined as M/
M, where M is the mass
being detected and M is the separation of the mass of
interest from an adjacent mass. The M can be measured at
the half-height of the mass peak or at 10% of the peak height,
depending on the inherent shape of the mass peak in the mass analyzer
(e.g., quadrupole mass filter vs magnetic/electric sector). A
quadrupole MS delivers unit mass resolution at half-height, with an R
of ~500 at m/z 500. With an R of 5000, one can
selectively detect ethanol (C2H4O,
m/z 46.042) in the presence of formic acid
(CH2O2, m/z 46.005) without
chromatographic separation. The advantage of GC/HRMS in the detection
of anabolic agents is shown in Fig. 2
, in which a urine sample supplemented with clenbuterol, 1
µg/L, was analyzed with the same GC/HRMS instrument but with
adjustment of the mass resolution. The use of GC/HRMS for routine
analysis of environmental dioxins has been required for a decade, and
its advantages for isotope-dilution methods for cortisol analysis were
recognized in the 1970s. The main disadvantages to GC/HRMS instruments
are their relatively large size, the considerably increased costs for
purchase and maintenance, and the degree of expertise required for
their operation.
|
A similar enhancement in LOD might be expected from tandem MS (MS/MS), in which monitoring a selected product ion produced from the collision-induced dissociation (CID) of a precursor ion can provide exquisite selectivity (5). Tandem MS can be performed either in space, with use of sequential sets of quadrupole rods, or in time, utilizing the storage capabilities of the ion trap. Using a Varian Associates (Palo Alto, CA) Saturn III QIT in the MS/MS mode and large-volume temperature-programed injection techniques, Borts and I have detected 100 fg of norandrosterone injected on-column, corresponding to a 6 ng/L concentration in urine (6). The LOD achieved for other steroids will depend on the availability of a relatively intense, high-mass precursor ion (6). Internal ionization ion traps (e.g., Varian Saturn III) and external ionization ion traps (e.g., Finnigan GCQ) appear to have some differences in LOD (J. Segura, personal communication). The application of GC/MS/MS technology to screening methodology will present some difficulties, because the sequential mass isolation and CID for large numbers of ions present the same sampling time limitations discussed above for SIM. It remains to be seen whether improved acquisition software can improve this situation.
Improved GC resolution could also improve the LOD, which is limited by the number of compounds extracted from the matrix. We postulated that the three-dimensional rigidity of the new generation of polysilarylene-bonded-phase capillary GC columns might improve selectivity between various steroid isomers; however, this hypothesis could not be confirmed (6). Similar lack of improved selectivity was observed for polychlorinated biphenyls/dioxins in environmental testing (M. Hastings, personal communication), suggesting that, when the polycyclic backbone of the compound of interest is sufficiently large, the phenyl moiety in the stationary phase cannot be closely approached and thus cannot impart stereochemical selectivity. Application of improved chromatographic techniques, e.g., multidimensional GC (7), could potentially be used with high-speed time-of-flight MS. Effective implementation of multidimensional GC requires the use of independent, orthogonal separation mechanisms (8). Although the above-mentioned phenylmethyl-type columns do not provide orthogonality, the use of cyanopropyl- or trifluoropropyl-bonded phases based on the polysilarylene approach should provide different selectivities while providing the higher temperature limits and robustness required for routine steroid analysis.
Improved detection limits could also be achieved with improved sample
clean-up strategies (see Fig. 1
). Immunoaffinity extraction and HPLC
preparation of steroids have been reported [9]
but at present are difficult to apply to the screening of large numbers
of samples. Detection of the residues of anabolic agent administration
for long times after the administration requires not only sensitive
analytical techniques and appropriate timing of specimen collection but
also selection of an appropriate long-lived metabolite
(3). For example, the chemical rearrangement of the
sulfate metabolites of 17
-methyl-17ß-hydroxy steroids to form
18-nor-17,17-dimethyl-13(14)-en compounds
(3)(10) may provide a selective approach to
the detection of this group of orally active anabolic steroids.
Although much progress has been made in detecting exogenous steroid
use, detecting the use of normally occurring steroids (e.g.,
testosterone, dihydrotestosterone, and dehydroepiandrosterone) has been
more difficult. A testosterone/epitestosterone (T/E) ratio >6 has been
used to detect testosterone abuse (11). Dehennin has
suggested that many of the cases of what has been termed
"naturally-elevated" T/E are attributable to the excretion of
substantial amounts of epitestosterone sulfate (12), a
compound that would not be detected in the usual steroid procedure. We
have developed a method for directly detecting the sulfate and
glucuronide conjugates of steroid metabolites by using HPLC/MS/MS
(13). Deuterated internal standards are used for both
conjugates of testosterone and epitestosterone (Fig. 3
) because of the effect of the interface and collision energy on
the resulting product ion
production.1
Four internal standards correct for any fluctuation in
formation of the ammonium adduct used for sulfate conjugate detection
and any fragmentation, as was observed for the glucuronide conjugates.
The results to date suggest that about one-third of the cases of
"naturally elevated" T/E are the result of unusually high
concentrations of epitestosterone sulfate and low concentrations of
epitestosterone glucuronide. In the course of our studies, we verified
that concentrations of testosterone glucuronide in the urine of Chinese
men are low, although the amounts of testosterone sulfate are similar
to those of Caucasian men.1
The use of direct detection of
the steroid conjugates with HPLC/MS/MS provides a powerful tool to
investigate stage II metabolism of steroids, and studies are underway
in our laboratory to (a) understand the racial differences
in testosterone metabolism and excretion and (b) identify
additional metabolites that indicate exogenous use.
|
Another promising approach to detecting the use of exogenous testosterone has been the use of GC/combustion/isotope ratio MS (14)(15). The premise of this approach is that the 13C abundance of a pharmaceutical preparation of testosterone is differs from that in testosterone produced from dietary precursors. A comparison of the 13C abundance of testosterone with that of its metabolic precursors that indicated a difference in abundance would support a finding of a nonmetabolic source of testosterone. Although preliminary results appear promising, much work must be done to validate this new methodology. In particular, the detection of exogenous use in the presence of a mixture of trace amounts of native and exogenous testosterone will be a challenge.
In our increasingly litigious society, athletes rely on such defense strategies as the effect of ethanol on the testosterone/epitestosterone ratio (16), inadvertent ingestion (17), and bacterial contamination to attempt to avoid the penalties for detection of banned substances in their urine. In most circumstances, the laboratory has methods at its disposal to refute these challenges. Nevertheless, improved LOD alone will not resolve these issues, and research is underway in several laboratories to provide the appropriate pharmacological foundation for further testing.
peptide hormones
The IOC banned substance list includes adrenocorticotropin, hGH,
chorionic gonadotropin (hCG), erythropoietin (EPO), and their releasing
factors. Although detection of these peptides by immunoassay is well
established in the clinical laboratory, the acceptability of this
method of analysis as definitive proof of administration has yet to be
established in forensic toxicology. Recent advances in the generation
of ions directly from liquids has suggested the possibility of
confirmation of peptide hormones by MS. The use of ES ionization allows
the production of multiply charged ions from peptides and proteins,
which extends the effective mass range of the mass spectrometer. Fig. 4
shows the ES mass spectrum of hGH from Genentech and Lilly. The
addition of a single methionine to the carboxy-terminal end of the
Genentech peptide allows a clear differentiation of the mass of the two
proteins, distinguishing between 22 124 Da and 22 255 Da. Thus,
detection of the recombinant hGH in the presence of native protein
would be relatively easy. The second issue for MS confirmation of
peptides in urine is the LOD. Little trace quantitative work of
proteins has been done with HPLC/MS. We applied three-ion SIM analysis
to the HPLC/MS detection of IGF-1. An HPLC/MS trace for injection of 10
fmol of IGF-1 is shown in Fig. 5
(18), establishing that detection limits with ES
HPLC/MS should not be a problem.
|
|
Determination of glycoproteins such as hCG and EPO presents additional challenges because of the presence of oligosaccharide moieties. The microheterogeneity of the oligosaccharides gives rise to a large number of molecules that all have the same amino acid sequence but different masses and charges. This precludes direct monitoring of the intact peptide by use of ES ionization. Monitoring the unique tryptic digest fragments provides an opportunity to identify peptide hormones. Both ES ionization HPLC/MS (19) and matrix-assisted laser desorption ionization–time-of-flight mass analysis ("MALDI-TOF") (20) have been applied to detection of hCG. Whether either or both techniques have the ability to produce quantitative as well as qualitative identification data remains to be established. Enrichment of peptide hormones from urine by immunoaffinity trapping can be readily achieved (21), which should facilitate trace determination of the peptide hormones. Differences in the electrophoretic mobility of native and recombinant EPO were used to detect administration of the latter in 11 of 15 urine samples (22). Work is currently underway in our laboratory to determine whether the observed charge differences can be exploited to develop an HPLC/MS method for detection of recombinant EPO.
Numerous issues surround which markers are necessary for identification of peptides. For example, the ions used for SIM analysis of IGF-1 mentioned above are multiply charged species formed by multimolecular reaction dynamics at atmospheric pressure. This is distinctly different from the electron ionization unimolecular decomposition reactions that are widely accepted as indicative of structural features in GC/MS analysis. The reproducibility of ion ratios for ES ionization must be established. In addition, the number of ions necessary to establish the "identity" of the peptide has not been tested, and the use of tryptic digests raises the issue of how many tryptic fragments are necessary to establish identity of a protein. Clearly, the combination of immunoaffinity extraction from urine with detection by MS increases selectivity, but the extent to which it should relax the above criteria must be determined. Although these analytical standards must be established in the laboratory, the legal admissibility of MS confirmation of peptides and proteins will be established by precedent.
In summary, application of state-of-the-art analytical tools to detection of performance-enhancing compounds will decrease the limits of detection and increase the numbers and types of compounds that can be detected. Analytical advances will not only benefit athletic drug testing, but should provide new methods for, e.g., characterizing protein standard materials (19). Pharmacokinetic and pharmacodynamic data must also be accumulated to support the analytical advances, so that fair and equitable medical review of cases can be undertaken. Given the amount of financial resources invested in sport, the athletes (and fans) have the right to competition decided by innate ability and hard work—not by potentially dangerous pharmacological intervention.
Footnotes
2 Nonstandard abbreviations: CID, collision-induced dissociation; EPO, erythropoietin; ES, electrospray; GC/MS, gas chromatography/mass spectrometry; hGH, human growth hormone; hCG, human chorionic gonadotropin; HRMS, high-resolution mass spectrometry; IGF-1, insulin-like growth factor-1; IOC, International Olympic Committee; LOD, limits of detection; MS/MS, tandem mass spectrometry; QIT, quadrupole ion trap; SIM, selected- ion monitoring. 
1 Sanaullah, Borts DJ, Bowers LD. Direct measurement of testosterone and epitestosterone conjugates in urine by HPLC/MS/MS. Manuscript in preparation.
References
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  C. M. McHugh, R. T. Park, P. H. Sonksen, and R. I.G. Holt Challenges in Detecting the Abuse of Growth Hormone in Sport Clin. Chem., September 1, 2005; 51(9): 1587 - 1593. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. M. Tokish, M. S. Kocher, and R. J. Hawkins Ergogenic Aids: A Review of Basic Science, Performance, Side Effects, and Status in Sports Am. J. Sports Med., September 1, 2004; 32(6): 1543 - 1553. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. A. Cowan and A. T. Kicman Doping in Sport: Misuse, Analytical Tests, and Legal Aspects Clin. Chem., July 1, 1997; 43(7): 1110 - 1113. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
