Detection of sibutramine administration: a gas chromatography/mass spectrometry study of the main urinary metabolites



A gas chromatographic/mass spectrometric (GC/MS) study aimed at identifying the metabolites of sibutramine (1-(4-chlorophenyl)-N,N-dimethyl-α-(2-methylpropyl)cyclobutanemethanamine) in urine is described. Urinary excretion of sibutramine metabolites following the oral administration of a single dose of sibutramine was followed by GC/MS analysis. After identification of the chromatographic signals corresponding to the six main urinary metabolites, the fragmentation pattern was studied in electron ionization (EI) mode after derivatization to the corresponding methyl and trimethylsilyl derivatives. Urine samples were pretreated according to a reference procedure (liquid/liquid separation, enzymatic hydrolysis, pre-concentration under a stream of nitrogen and derivatization, either under thermal incubation and by microwave irradiation). All sibutramine metabolites were excreted as glucuroconjugates, and retain the chiral carbon present in the sibutramine skeleton. The metabolites identified included mono-desmethylsibutramine (nor-sibutramine), bi-desmethylsibutramine (nor-nor-sibutramine), and the corresponding hydroxylated compounds, the hydroxylation taking place either on the cyclobutane or on the isopropyl chain. The excretion profiles of the different metabolites were also evaluated. From an analytical point of view, the method can be applied to different fields of forensic analytical toxicology, including anti-doping analysis. Although the lack of certified reference materials does not allow a precise determination of the limits of detection (LODs) of all the sibutramine metabolites, an estimation taking into account the response factor of similar compounds ensures that all metabolites are still clearly detectable in a range of concentrations between 10 and 50 ng/mL, thus satisfying the minimum required performance limits (MRPLs) of the World Anti-Doping Agency (WADA). Copyright © 2006 John Wiley & Sons, Ltd.

Sibutramine (1-(4-chlorophenyl)-N,N-dimethyl-α-(2-methylpropyl)cyclobutanemethanamine; for chemical structure, see Fig. 1) is a relatively new drug approved for the management of obesity. The mechanism of action involves the inhibition of serotonin (5-hydroxytryptamine) and norepinephrine reuptake (for a general review of the main pharmacological features, see Refs. 1 and 2). However, since this effect is exerted by sibutramine in vivo but not in vitro, it is likely that the parent compound act merely as a pro-drug, the active principle(s) being one or more of its metabolites.3

Figure 1.

Molecular structure of 1-(4-chlorophenyl)-N,N-dimethyl-α-(2-methylpropyl)cyclobutanemethanamine (sibutramine). * asymmetric carbon.

Sibutramine undergoes primary and secondary metabolism,4, 5 and its elimination occurs after biotransformation to at least six metabolites. More specifically, sibutramine is metabolized in the liver to form the corresponding demethylated metabolites, which are in turn hydroxylated, and finally conjugated. Although no final information is available, it is likely that only the glucuroconjugated metabolites are excreted in appreciable amounts in urine, while the others are eliminated via the entero-hepatic route. To date, despite the increasing use of sibutramine and a growing set of information concerning its pharmacological effects, there are not sufficient data on either the characterization of sibutramine metabolites or their determination in biological fluids and tissues.

The use of sibutramine by athletes is forbidden, effective January 2006, since this drug is now put together with other similar drugs (e.g. dexfenfluramine) in the class 'S6' (stimulants) of the World Anti-Doping Agency (WADA) list of prohibited substances and methods.6 It is therefore necessary for all accredited anti-doping laboratories to screen all urine samples collected 'in competition' for the presence of sibutramine. A method for the determination of some sibutramine metabolites by liquid chromatography/tandem mass spectrometry has recently been presented.7 To date, no methods for the detection of sibutramine and/or sibutramine metabolites by gas-chromatography/mass spectrometry (GC/MS) have been described.

The present paper proposes a GC/MS method for the determination of the urinary metabolites of sibutramine. An urinary excretion study has been carried out by collecting urine samples for three days after oral administration of a single dose (10 mg) of sibutramine. Following identification of the chromatographic signals corresponding to the main urinary metabolites of sibutramine, the metabolite fragmentation patterns were studied in electron ionization (EI) mode after derivatization to the corresponding methyl and trimethylsilyl derivatives.

The optimal GC/MS conditions for the detection of the metabolites were also studied, taking into account the general requirements of a routine anti-doping analysis.


Chemicals and reagents

Ammonium iodide, dithioerythritol (DTE), N-methyl-bis(trifluoroacetamide) (MBTFA) and methyltestosterone (used as internal standard, ISTD in the GC/MS study of the TMS derivatives) were supplied by Sigma-Aldrich (Milano, Italy). All other reagents, including methyl iodide, were supplied by Carlo Erba (Milano, Italy); N-methyl-N-(trimethylsilyl)trifluoroacetamide (MSTFA) was obtained from Macherey-Nagel (Düren, Germany). The derivatizing agents were for methylation a mixture of methyl iodide/acetone 1:10 and for silylation a mixture of MSTFA/NH4I/DTE (1000:4:2). Enzymatic hydrolysis was carried out using β-glucuronidase from E. coli supplied by Roche (Mannheim, Germany); sibutramine and mono-desmethylsibutramine standards were obtained from Suntech (Hangzhou, Zhejiang, China) and Synfine (Richmond Hill, Ontario, Canada), respectively. Reductil (sibutramine·HCl·H2O, 10 mg) was purchased from Knoll Deutschland GmbH (Ludwigshafen, Germany).

GC/MS conditions

The GC/MS system was an Agilent HP6890 (Agilent Technologies Italia, Rome, Italy) gas chromatograph coupled to a 5973 mass detector operated at 70 eV in EI mode. The chromatographic conditions were the following: (i) analysis of trimethylsilyl (TMS) derivatives: J&W methylsilicone capillary column (17 m × 0.2 mm i.d., 0.11 µm film thickness); the oven temperature was held at 185°C for 2.5 min, increased to 210 at 3°C/min, then increased to 290°C at 15°C/min, and held for 3 min. (ii) analysis of methyl derivatives: J&W 5% phenyl-methylsilicone capillary column (17 m × 0.2 mm i.d., 0.33 µm film thickness); the oven temperature was held at 90°C for 1.5 min, increased to 260°C at 25°C/min, held for 3 min, and increased to 300°C at 25°C/min (held for 3 min). In both cases the injection port was set at 280°C in split mode (10:1) and helium was used as carrier gas at a constant pressure of 18 psi.

The mass detector was operated in scan mode (scan range from m/z 50 to 450) and in selected ion monitoring (SIM) mode for the excretion profiling of the TMS derivatives of sibutramine metabolites (see Table 1).

Table 1. Main characteristic ions, acquired in SIM mode, for the substances considered in the present study (ion used for quantitation is given in bold)
Target compoundDiagnostic ions [m/z]
nor-nor-S TMS158 102 115 128 137
nor-S TMS172 102 129 115 137
OH(isopropyl chain)-nor-nor-S 2TMS246 156 340 266 102 115
OH(isopropyl chain)-nor-S 2TMS260 170 129 156 410 412
OH(cyclobutane)-nor-nor-S 2TMS158 102 115 129 238 240 264 396
OH(cyclobutane)-nor-S 2TMS172 129 114 156 410 114
ISTD301 446

Sample preparation

Urine samples were analyzed first as free fractions and, secondly, after hydrolysis with β-glucuronidase and derivatization to the methyl and TMS derivatives. In the case of the free fraction, 50 µL of the ISTD (methyltestosterone 1 µg/mL or indometacine 20 µg/mL for methyl derivatives), 2 mL of carbonate buffer (pH 9) and 10 mL of tert-butyl methyl ether were added to 3 mL of urine samples and liquid/liquid extracted for 10 min. The ether phase was brought to dryness and derivatized to form either the TMS or the methyl derivatives, respectively (i) by addition of 50 µL of MSTFA/NH4I/DTE and incubation at 70°C for 30 min, or (ii) by addition of 200 µL of acetone/methyl iodide and 50 mg of potassium carbonate and incubation at 110°C for 10 min (or irradiated by microwaves at 1200 W emitted energy for 6 min). Additional experiments carried out to clarify the structure of the hydroxylated sibutramine metabolites were carried out on trifluoroacetyl (TFA) derivatives, obtained by incubating the dried ether extract with 50 µL of MBTFA at 70°C for 30 min. For the analysis of conjugated metabolites, prior to the described procedures the samples were treated with 1.5 mL of phosphate buffer (pH 7.4) and 30 µL of β-glucuronidase and incubated at 50°C for 1 h.

Urinary excretion study of sibutramine metabolites

An excretion study was performed on two healthy female volunteers (age: 28 and 29) who gave informed consent prior to the study. A blank urine was collected immediately prior to the administration of a tablet of Reductil (10 mg sibutramine hydrochloride hydrate, corresponding to 8.37 mg of free sibutramine). Urine samples were then collected on the following three days. Relative concentrations were estimated by comparing the area ratios of the most abundant ion of each compound with that of the ISTD; all values were then corrected for specific gravity.


Metabolite identification

GC/MS analysis of the urine samples collected for three days after oral administration of sibutramine confirmed that its elimination occurs after extensive biotransformation. No sibutramine was detected in the urine samples. In addition, no significant amounts of sibutramine metabolites are excreted in the urine free fraction (i.e. as non-conjugates), and this finding could also explain why sibutramine itself, lacking conjugation sites in its molecular structure, is not excreted in urine. Consequently, we have focused our attention on glucuroconjugates, and analyzed urine samples after enzymatic hydrolysis by β-glucuronidase. The resulting extract was then derivatized either by methyl iodide or by MSTFA/NH4I/DTE, to form the corresponding methyl and TMS derivatives, respectively. The complete picture resulting from this investigation shows that sibutramine is converted into mono- and bi-desmethyl metabolites, which are in turn hydroxylated, either on the cyclobutane ring or on the isopropyl chain. Since all these compounds (i) retain the chiral centre originally present on the sibutramine structure (this finding being in agreement with in vitro data obtained in primary cultures of rat hepatocytes5), and (ii) undergo glucuroconjugation, a total of 12 glucuronides (6 pairs of enantiomers) can in principle be formed and excreted in urine.

When the samples were analyzed, after enzymatic hydrolysis, as the methyl derivatives (thus converting the mono- and bi-demethylated metabolites back into sibutramine), a considerable amount of sibutramine was indeed detected. This confirms that all the metabolites are de-methylated.

In order to achieve better sensitivity and to gain more structural information on the nature of the metabolites, the samples were also analyzed as TMS derivatives. After deconjugation and trimethylsilylation the chromatographic pattern shown in Fig. 2 was obtained.

Figure 2.

Representative chromatographic pattern of sibutramine metabolites in urine; data refer to a urine sample collected 2 h after oral administration.

The fragmentation study suggested the presence of six major metabolites, as shown in Figs. 3–8, these being: mono-desmethylsibutramine (nor-S), bi-desmethylsibutramine (nor-nor-S), hydroxylated mono-desmethylsibutramine, with the oxydryl group linked to the cyclobutane ring (OH-nor-S1), hydroxylated mono-desmethylsibutramine, with the oxydryl group linked to the isopropyl chain (OH-nor-S2), cyclobutane-hydroxylated bi-desmethylsibutramine (OH-nor-nor-S1), and isopropyl chain hydroxylated bi-desmethylsibutramine (OH-nor-nor-S2), all of them excreted as glucuronides. This picture was consistent with data obtained on both the TMS and the TFA derivatives. No hydroxylation occurred on the aromatic ring, as confirmed by a series of experiments carried out by performing the extraction step at pH 14: the resulting chromatographic pattern presented no difference from the one obtained after extraction at pH 9, thus showing that no phenolic hydroxy groups were present in any of the studied molecules. Finally, unchanged sibutramine was not detected in any of the urine samples.

Figure 3.

nor-S TMS: (a) mass spectrum and (b) fragmentation study.

Figure 4.

nor-nor-S TMS: (a) mass spectrum and (b) fragmentation study.

Figure 5.

OH(isopropyl chain)-nor-S 2TMS: (a) mass spectrum and (b) fragmentation study.

Figure 6.

OH(isopropyl chain)-nor-nor-S 2TMS: (a) mass spectrum and (b) fragmentation study.

Figure 7.

OH(cyclobutane)-nor-S 2TMS: (a) mass spectrum and (b) fragmentation study.

Figure 8.

OH(cyclobutane)-nor-nor-S 2TMS: (a) mass spectrum and (b) fragmentation study.

The trend of the excretion of the main metabolites is shown in Fig. 9. All the metabolites present the same excretion tendency, with hydroxylated metabolites excreted to a major extent, and with particularly OH-nor-nor-S1 detectable for a longer time and with a higher response.

Figure 9.

Excretion pattern of the main sibutramine metabolites. The Y axis indicates the values of the area ratio between the target compound and the ISTD. ▪ = OH(cyclobutane)-nor-nor-S; ○ = OH(isopropyl chain)-nor-nor-S; ▵ = OH(cyclobutane)-nor-S; □ = OH(isopropyl chain)-nor-S.

Some considerations arise from the data presented here that are relevant both to sibutramine metabolism and to the possibility of detecting sibutramine administration:

  • 1All compounds considered in this study can be effectively derivatized to form the corresponding trimethylsilyl, methyl, and trifluoroacetyl derivatives;
  • 2unchanged sibutramine is not detectable in urine, at a sensitivity of 10 ng/mL;
  • 3sibutramine is eliminated in urine after conversion into at least six metabolites/pairs of metabolites, all glucuroconjugates: mono-desmethyl, bi-desmethyl, hydroxylated monodesmethyl and hydroxylated bidesmethylsibutramine (with the hydroxylation taking place on either on the cyclobutane or on the isopropyl chain);
  • 4based on the fragmentation patterns obtained in the present study, sibutramine administration can be detected by the urinary analysis of one or more of their metabolites, either by methylation, silylation, or trifluoroacetylation, but only after hydrolysis of the glucuronides;
  • 5the limits of detection vary in the range 10–50 ng/mL, thus matching the minimum required performance limit (MRPL) for stimulants, fixed by the WADA at 500 ng/mL;8
  • 6the window of detectabilty is longer than 48 h for all the above metabolites, the maximum being for hydroxylated metabolites, in particular OH(cyclobutane)-nor-nor-S, that is still detectable (again at a sensitivity of 10 ng/mL) 60 h after administration: these data are more than sufficient to allow the detection of sibutramine administration in anti-doping analysis, as sibutramine is included in the class of stimulants and therefore forbidden only 'in competition'.


This work was supported in part by a research grant from the Italian Department of Health – National Antidoping Commission ('Commissione di Vigilanza sul Doping del Ministero della Salute'). The authors also wish to express their gratitude to Francesca De Angelis for her valuable technical support.