1H NMR Urine Analysis as an Effective Tool to Detect Creatine Supplementation

Transcription

1 ] TechnicalNotel 1H NMR Urine Analysis as an Effective Tool to Detect Creatine Supplementation Bernard Cartigny 1, Nathalie AzarouaP, Laurence Mille-Hamard 3, Michel Imbenotte 4, Pascal Kintz s, Gaston Vermeersch 2, and Michel Lhermitte 1,4,* 1Laboratoire de Biochimie et Biologie Mol~culaire, H6pital Calmette, Lille, France; 2Laboratoire de Physique, UPRESA CNRS 8009, Laboratoire d'application RMN de I'Universit~ de Lille 2, BP 83, Lille, France; 3Laboratoire d'etucles de la Motricit~ Hurnaine, Universitd de Lille 2, Lille, France; 4Laboratoire de Toxicologie, Facult~ des Sciences Pharmaceutiques et Biologiques, BP 83, Lille, France; and 51nstitut de Mddecine L~gale, I 1 rue Humann, Strasbourg, France Abstract I Creatine is one of the main compounds in muscular energetic metabolism leading to phosphocreatine to maintain high ATP levels. Creatine is found in blood and excreted in small amounts in urine. Creatine supplementation and athletic performances are supposed to be correlated, particularly in intensive and intermittent efforts. After oral creatine supplementation, a 1H nuclear magnetic resonance (NMR) spectroscopy method was developed for its direct analysis, without any pretreament of urine samples. This method can be used to detect any supplementation of creatine, a substance prohibited in France. The detection limit is 10 pmol/l (1.31 mg/l) and analysis is performed in 10 min. After a single oral supplementation of 2.1 g to three subjects, a kinetic investigation reveals a maximum concentration of 20 mmol/l (2.62 g/l), observed between 1 and 6 h after ingestion. This procedure was used to test 13 urine specimens obtained from bodybuilders. From the concentrations measured (range: 0.41 to retool/l, 54 to 1350 rag/l), the doping practices of at least nine athletes could be observed. Creatine is not often analyzed in hospital laboratories. This paper documents how easily creatine can be determined and quantitated by 1H NMR spectroscopy. Introduction Approximately 95% of total creatine is located in skeletal muscles in free and phosphorylated forms, the latter being essentially used in intense short-term activities (1). The phos- * Author to whom correspondence should be adressed: Pr. Michel Lhermilte Laboratoire de Bioehimie et Biologie Mo16culaire, H6pital Calmelte, Lille, France. mlhermitle@chru-lille.ff. phorylation step is reversible and phosphocreatine maintains the adenosine triphosphate/adenosine diphosphate ratio in muscle tissues (2). Creatine liberated from phosphocreatine undergoes nonenzymatic cyclization to form creatinine, which is excreted in urine in high amounts and at a constant rate (3). Therefore, creatine may be used as a dietary supplement, and a positive effect of pure creatine ingestion has sometimes been noted on physical performances (4-6). Creatine is now commercially available over-the-counter in the United States and is widely used by athletes. The daily ingested dose is generally 5 g, which is equivalent to about I kg of uncooked meat (7). On the contrary, the sale of creatine is prohibited in France; it is considered a doping substance. Most published analytical techniques for creatine analysis involve high-performance liquid chromatography (8,9). With liquid chromatography coupled to mass spectrometry in the chemical ionization mode, Yasuda et al. (3) determined creatine, creatinine, and glycocyamine in serum and urine. More recently, Burke et al. (10) separated and quantitated creatine and creatinine by capillary electrophoresis with ultraviolet (UV) detection. They followed the excretion of creatine in urine as a function of time, after creatine ingestion by two athletes (100 mg/kg body mass). Following ingestion, the maximum of creatine excretion in urine is after about 2-3 h and a return to basal values is noted after 13 to 14 h. In physiological conditions, reference values of creatine excretion are scarce. According to Bales et al. (11), creatine excretion can rise to 200 mmol/mol creatinine (26.2 g/mole creatinine). In the paper by Yasuda et al. (3), its mean excretion is mmol/l (24.9 mg/l) in humans. According to Beyer (12), creatine excretion in adults ranges from 150 to 1200 pmol/24 h (20 to 157 rag/24 h). Reproduction (photocopying) of editorial content of this journal is prohibited without publisher's permission. 355

2 ]H Nuclear magnetic resonance (NMR) analysis can be used to determine a wide range of small molecules in biological fluids (11) and offers several advantages: it is rapid, nondestructive, and the determination of many different chemical species is possible on the same analysis. This technique has already contributed to clinical diagnosis (13,14) and thus has a promising use in urine analysis. Among compounds which can be detected, creatinine determination by 1H NMR spectroscopy was already tested by Chinayon et al. (15). Considering the chemical analogy between creatine and creatinine and their metabolic link, it is suitable to determine both compounds simultaneously. Their detection and quantitation can be obtained by 1H NMR spectroscopy using their respective singlet peaks at characteristic chemical shifts. In this paper, the applicability of 1H NMR spectroscopy to detect creatine supplementation was evaluated on spiked control urines. After the validation step, the method was applied to a kinetic investigation of creatine excretion by three healthy volunteers, over 48 h. As proof, creatine supplementation was creatlnine CH 3 crea6ne A... I" "P" '1' '" I",'1",,q,,,,I,,,,1,,,,1,,,,I,,,,I,,,,I,,,,I,,,,I,,,,i,,,,i,,,,i,,,,i,,,,i,,, / i I creatinine J creatine B... i,,,,i,,,,i,,,,l,,,,i,,,,i,,,,i,,,,~,,,,i,,,,i,,,,i,,,,i,,,,i,,,,i,,,,i,,,,i,,,,i,,,,i,,,,i,,, ,5 0.0 creat bine /\ C Figure 1.1H NMR spectrum (300 MHz) of urine ph 5.0, during creatine loading test: To (A); To + 6 h after ingestion (B); and To + 36 h after ingestion (C). Table I. Urinary Creatine and Creatinine Concentrations from People Having Preferential Consumption of Meat (cases 1 to 4) or Fish (cases 5 to 9) I Table II. Urinary Creatine and Creatinine Concentrations from 13 Athletes Who Did Not Ingest Creatine I Table III. Urinary Creatine and Creatinine Concentrations from 13 Bodybuilders Who Ingested Creatine Supplement

3 investigated on urine samples obtained from 13 bodybuilders, suspected of ingesting this compound. Material and Methods Apparatus and chemicals 1H NMR spectra were recorded on a Bruker DPX 300 MHz spectrometer (Bruker S.A., Wissembourg, France) at ambient probe temperature. Colorimetric measurements of creatinine using Jaffd reaction were performed on an Olympus AU 600 automat (Olympus, Rungis, France). 3-Trimethylsilyl 2,2',3,3'-tetradeuteropropionic acid (TSP-d4) and deuterium oxide were purchased from Eurisotop (Saint Aubin, France). Other standard compounds, creatine, creatinine, and glycine, were obtained from Sigma Aldrich (Saint Quentin Fallavier, France). The sale of creatine for supplementation is not allowed in France; therefore, capsules containing 700 mg of creatine monohydrate were purchased in the U.S. (Twin Laboratories, Hauppauge, NY). Analytical methods The urine samples collected were directly used for 1H NMR analysis. A 500-1JL sample (urine or standard) was ajusted to ph 5.0 and introduced into a 5-mm diameter NMR tube. A titrated solution of TSP-d4 in deuterium oxide was added to a capillary tube that was coaxially inserted into the NMR tube. This provided an internal field-frequency lock and was used as reference for chemical shift (5 = 0.00 ). A presaturation sequence was used to suppress the intense water signal. De- pending on the sample concentration, 128 to 512 transients were collected into a 16 K data point computer, with a spectral width of 3200 Hz and a 30 ~ pulse. Prior to Fourier transform, an exponential apodization function was applied, corresponding to a 0.3 Hz broadening of the line. Data processing was carried out using the 1D WIN NMR program from Bruker. Calibration procedure: creatine or creatinine were spiked in control urines by known amounts in the range 1-20 mmol/l ( mg/l). All quantitation processes were run on methyl groups of creatine and creatinine by relative integration of their peak to the signal of TSP-d4 peak. The investigations were performed on urine samples from volunteers and three groups were considered. First, to measure normal values of creatine in different conditions, four urine specimens were obtained from persons with a preferential consumption of meat and five urine specimens from persons with a preferential consumption of fish. With the same purpose, urine specimens obtained from 13 athletes not having ingested creatine were also tested. Secondly, to follow the excretion of creatine in urine, three healthy volunteers ingested 2.1 g of creatine as a single oral supplementation. Samples were collected 1, 3, 6, 12, 24, 36, and 48 h after ingestion. Thirdly, 13 urine specimens were obtained from bodybuilders considered by the French customs as responsible for trafficking doping agents. Several hundred tablets and ampoules of various anabolic steroids were discovered at their homes or at their gymnasium. Moreover, several kilograms of creatine were also found. To document the doping practices of these bodybuilders, the judge in charge of the case requested urine analyses. For each group, all urine samples were adjusted to ph 5.0, immediately frozen, and stored at-20~ until analysis O IS O 10 ~ m 9 m m 9 m -- To To+lh T0+3h To+6h To+12h To+24h Time after ingestion (To + hours) m m To + 36 h To + 48 h Figure 2. Evolution of urinary creatine excretion after creatine loading test in three healthy volunteers. 357

4 Results and Discussion By 1H NMR, creatine is characterized by two singlets, one at 3.03 corresponding to the methyl group and the other one at 3.92 attributed to the methylene group. Creatinine, which is a creatine metabolite, is excreted at a constant rate and is always present on urine spectra (Figure 1). It is also characterized by two singlets at 3.03 and 4.17 (CH3 and CH2), and the methyl groups of the two compounds are not well separated from each other. But the resonance frequency of creatine signals are independent of ph values, whereas creatinine signals vary slightly with ph values. Adjustment of the ph of the samples to 5.0 makes it easy to separate the methyl groups of creatine and creatinine (3.03 and 3.08, respectively). For control urine samples obtained from laboratory personnel, spiked with creatine, and corrected for endogenous creatine, a significant linear relationship was observed between experimental measurements of creatine (y) and theoretical values (x): y = x with a correlation coefficient r = When quantitation is performed on the methyl group, the detection limit is estimated at 10 IJmol/L (1.31 rag/l) and the quantitative limit is found to be 30 l~mol/l (3.93 mg/l). Urinary creatine concentrations for people eating meat or fish are reported in Table I. These are in the range of 0.21 to 0.53 mmol/l (27.5 to 69.5 mg/l) after consumption of meat and in the range of 0.19 to 1.18 mmol/l (24.9 to mg/l) after consumption of fish. Table II presents values found in athletes who had not ingested creatine. These are from 0.05 to 0.76 mol/l (6.6 to 99.7 rag/l). The values presented in Tables I and II, expressed as retool/tool creatinine, are quite similar and in agreement with the results of Bales et al. (11), which have showed that urinary creatine varied from 0 to 200 mmol/mol creatinine (with one value at 654). After creatine administration experiments on three healthy volunteers, urine analyses showed that the majority of creatine is rapidly excreted. Figure 1 shows three IH NMR spectra of urine samples during a creatine loading test. Creatine, normally present in urine at To ( mmol/l, mg/l), rapidly increases to 19 mmol/l (2.49 g/l) in the first hours after ingestion and progressively decreases to normal values after 24 h. Data collected from the three subjects is presented in Figure 2 where creatine excretion is reported as a function of time. This confirms that the excretion is rapid because the maximal concentration is observed between I and 6 h. For the three subjects, the decrease in creatine to physiological concentrations (about 0.4 mmol/l, 52 rag/l) is effective after about 24 h. During supplementation tests and from a mechanistic point of view, other compounds occuring in creatine metabolism were investigated. Creatinine was simultaneously measured by 1H NMR and by Jaff~ reaction, and the results were very similar: variations in results did not exceed 12%. Glycine is also present in normal urine and can often be seen on 1H NMR urine spectra. In our experiment, creatine supplementation did not seem to have any influence on the excretion of creatinine, as was already noted by Ropero-Miller et al. (7). Glycine, sarcosine, and guanidino acetic acid were also investigated by 1H NMR. Glycine was not elevated by creatine supplementation, and the other two compounds could not be detected in urine analyses during the study. Quantitative results of the bodybuilders' urine analyses are reported in Table III. According to normal values published by Bales et al. (11) (0 to 200 mmol/mole creatinine), creatine concentrations can be considered as elevated for nine of them. According to Yasuda et al.'s (3) normal values (mean 0.19 retool/l, 24.9 rag/l) and to data presented Tables I and II, creatine concentrations can be considered as elevated for all of them. When presented, these results were not denied by any of the bodybuilders. Among these results, the 10th case had a very high creatine concentration due to a very low and unexplainable creatinine content; as urinary specific gravity was normal (1.017), this sample had obviously not been adulterated. Conclusions Creatine is generally not analyzed in hospital laboratories because of analytical difficulties. This work documents that creatine can easily be determined and quantitated by 1H NMR spectroscopy without any pretreatment of urine samples, except for ph adjustment. Creatine supplementation by three healthy volunteers shows that elimination is rapid, as no elevated creatine concentration can be observed 24 h after ingestion. Two singlets of creatine are specific to this compound. Even when it is ingested in association with other stimulants or pharmaceuticals, creatine can be determined and quantitated by 1H NMR spectroscopy. The application of the technique to the measuring of creatine in 13 urine samples from bodybuilders confirms that 1H NMR spectroscopy is applicable to the detection of creatine among subjects suspected of doping practices. Acknowledgments The authors are grateful to Mrs. Alexandra Tavernier, M.A. (University of Glasgow) and Professeur Agr~g~ (English), for her expert advice in the revision of the English manuscript. References 1. RD. Balsom, K. SOderlund, and B. Ekblom. Creatine in humans with special reference to creatine supplementation. Sports Med. 18: (1994). 2. S.P. Bessman and P.J. Geiger. Transport of energy in muscle. Science 24: (1981 ). 3. M. Yasuda, K. Sugahara, J. Zhang, T. Ageta, K. Nakayama, T. Shuin, and H. Kodama. Simultaneous determination of creatinine, creatine, and guanidinoacetic acid in human serum and urine using liquid chromatography-atmospheric pressure chemical ionization mass spectrometry. Anal. Biochem. 253: (1997). 358

5 4. H.B. Rossiter, E.R. Cannell, and P.M. Jakeman. The effect of oral creatine supplementation on the 1000-m performance of competitive rowers. J. Sports ScL 14: (1996). 5. C.N. Maganaris and R.J. Maughan. Creatine supplementation enhances maximum isometric force and endurance capacity in resistance trained men. Acta Physiol. Scand. 163: (1998). 6. I. Mujika, S. Padilla, J. lbanez, M. Izquierdo, and E. Gorostiaga. Creatine supplementation and sprint performance in soccer players. Med. Sci. Sports Exerc. 32: (2000). 7. J.D. Ropero-Miller, H. Paget-Wilkes, P.L. Doering, and B.A. Goldberger. Effect of oral creatine supplementation on random urine creatinine, ph, and specific gravity measurements. Clin. Chem. 46: (2000). 8. M. Dunnett, R.C. Harris, and C.E. Orme. Reverse-phase ionpairing high-performance liquid chromatography of phosphocreatine, creatine and creatinine in equine muscle. Scand. J. Clin. Lab. Invest. 51: (1991). 9. Y.D. Yang. Simultaneous determination of creatine, uric acid, creatinine and hippuric acid in urine by high performance liquid chromatography. Biomed. Chromatogr. 12:47-49 (1998). 10. D.J. Burke, P.G. MacLean, R.A. Walker, P.J. Dewar, and T. Smith Palmer. Analysis of creatine and creatinine in urine by capillary electrophoresis. J. Chromatogr. B Biomed. Sci. Appl. 732: (1999). 11. J.R. Bales, D.P. Higham, 1. Howe, J.K. Nicholson, and P.J. Sadler. Use of high-resolution proton nuclear magnetic resonance spectroscopy for rapid multi-component analysis of urine. Clin. Chem. 30' (1984). 12. C. Beyer. Creatine measurement in serum and urine with an automated enzymatic method. Clin. Chem. 39: (1993). 13. S. Maschke, A. Wahl, N. Azaroual, V. Crunelle, M. Imbenotte, M. Foulard, G. Vermeersch, and M. Lhermitte. 1H NMR urinalysis for detecting fish-odour syndrome. Clin. Chim. Acta 263: (1997). 14. A. Wahl, N. Azaroual, M. lmbenotte, D. Mathieu, G. Forzy, B. Cartigny, G. Vermeersch, and M. Lhermitte. Poisoning with methanol and ethylene glycol: 1H NMR spectroscopy as an effective clinical tool for diagnosis and quantification. Toxicology 128:73-81 (1998). 15. S. Chinayon, W. Jinsart, P. Pansin, S. Eiam Ong, S. Sitprija, and V. Sitprija. Identification of urinary metabolites and quantitative measurement of creatinine by a proton nuclear magnetic resonance spectrometry. J. Med. Assoc. Thai. 73: (1990). Manuscript received October 1,2001; revision received March 20,