Elsevier

Journal of Chromatography B

Volume 857, Issue 1, 15 September 2007, Pages 67-75
Journal of Chromatography B

Stereoselective analysis of bupropion and hydroxybupropion in human plasma and urine by LC/MS/MS

https://doi.org/10.1016/j.jchromb.2007.07.007Get rights and content

Abstract

A sensitive, stereoselective assay using solid phase extraction and LC–MS–MS was developed and validated for the analysis of (R)- and (S)-bupropion and its major metabolite (R,R)- and (S,S)-hydroxybupropion in human plasma and urine. Plasma or glucuronidase-hydrolyzed urine was acidified, then extracted using a Waters Oasis MCX solid phase 96-well plate. HPLC separation used an α1-acid glycoprotein column, a gradient mobile phase of methanol and aqueous ammonium formate, and analytes were detected by electrospray ionization and multiple reaction monitoring with an API 4000 Qtrap. The assay was linear in plasma from 0.5 to 200 ng/ml and 2.5 to 1000 ng/ml in each bupropion and hydroxybupropion enantiomer, respectively. The assay was linear in urine from 5 to 2000 ng/ml and 25 to 10,000 ng/ml in each bupropion and hydroxybupropion enantiomer, respectively. Intra- and inter-day accuracy was >98% and intra- and inter-day coefficients of variations were less than 10% for all analytes and concentrations. The assay was applied to a subject dosed with racemic bupropion. The predominant enantiomers in both urine and plasma were (R)-bupropion and (R,R)-hydroxybupropion. This is the first LC–MS/MS assay to analyze the enantiomers of both bupropion and hydroxybupropion in plasma and urine.

Introduction

Bupropion is commonly used for the treatment of depression or smoking cessation and has been studied for a number of other uses such as bipolar disorder, attention-deficit hyperactivity disorder and weight loss [1], [2], [3]. Bupropion is extensively metabolized, with <0.5% reported to be recovered intact in urine [4]. Three active metabolites have been identified: hydroxybupropion, resulting from cytochrome P450-catalyzed oxidation, and erythrohydrobupropion and threohydrobupropion, reduced diastereomers formed via non-P450-dependent pathways. Although all three metabolites are pharmacologically active, their potency and toxicity relative to bupropion have not been fully characterized. According to in vivo studies using the mouse anti-tetrabenazine model of depression, the activity of hydroxybupropion is approximately 50% of the parent drug, and the activity of both erythrohydrobupropion and threohydrobupropion is approximately 20% of bupropion [5], [6], [7], [8], [9]. The average elimination half-lives are 8, 19, 35 and 19 h for bupropion, hydroxybupropion, erythrohydrobupropion and threohydrobupropion, respectively [10].

Bupropion is formulated and used clinically as a racemic mixture. The enantiomers have been shown to rapidly racemize in phosphate buffer (pH 7.4, 25 °C): 42%, 62%, and >94% in 2, 4, and 24 h, respectively [11]. When hydroxylated on the N-t-butyl carbon, bupropion rapidly undergoes ring closure, forming hydroxybupropion with two chiral centers (Fig. 1). However, only the (R,R)- and (S,S)-hydroxybupropion diastereomers are observed, possibly because the (R,S)- and (S,R)-diastereomers are sterically hindered [11]. Hydroxybupropion also racemizes in phosphate buffer (pH 7.4, 25 °C), albeit much more slowly: approximately 0.1% and 2% in 2 and 24 h, respectively [11]. Due to this slower rate of racemization, plasma concentrations of hydroxybupropion enantiomers accurately reflect stereospecific formation and elimination without confounding by racemization.

Stereochemistry may be important for bupropion therapeutic effects. The potency of bupropion enantiomers in vitro does not differ from that of the racemate, although racemization at physiological conditions may explain the lack of stereoselectivity [12]. Racemic hydroxybupropion potency is similar to that of racemic bupropion, but (S,S)-hydroxybupropion is significantly more potent than both (R,R)-hydroxybupropion and racemic bupropion in a mouse depression model and in antagonism of acute nicotine effects in mice [13].

The primary enzyme in bupropion metabolism to hydroxybupropion is cytochrome P4502B6 (CYP2B6) [14], [15], [16], [17]. Racemic bupropion hydroxylation has been used as an in vitro probe for the activity of CYP2B6 in liver microsomes [18], [19], [20], [21], [22], [23], [24]. More recently, racemic bupropion clearance has been used as an in vivo probe for the activity of hepatic CYP2B6, and for assessing the effects of CYP2B6 single nucleotide polymorphisms and CYP2B6 drug interactions [25], [26], [27], [28], [29], [30], [31], [32]. We recently found that bupropion hydroxylation catalyzed by CYP2B6 and human liver microsomes is stereoselective, with the rate of (S)-bupropion hydroxylation about three times that of (R)-bupropion (submitted). Clearance of the bupropion enantiomers may provide a better in vivo probe for CYP2B6 activity. Nevertheless, there is no information regarding plasma concentrations of bupropion enantiomers.

The use of bupropion as an in vivo CYP2B6 probe is not straightforward. The non-stereoselective assays used to date have shown that the half-life of hydroxybupropion exceeds that of bupropion, suggesting that hydroxybupropion concentrations are elimination rate-limited rather than formation rate-limited. Therefore, the ratio of hydroxybupropion/bupropion area under the plasma concentration versus time curves does not accurately reflect the rate of bupropion hydroxylation in vivo. An alternative potential for assessing bupropion hydroxylation, and CYP2B6 activity, is the urine formation clearance of hydroxybupropion. Nonetheless, there is no information on urine hydroxybupropion formation clearance, or that of the enantiomers.

Several methods have been published for the stereoselective analysis of bupropion in plasma using coupled achiral–chiral liquid chromatography, an ovomucoid column, or an α1-acid glycoprotein column [33], [34], [35]. Recently, a method was published for the stereoselective analysis of hydroxybupropion in plasma using a Cyclobond I 2000 column [36]. All of these methods, however, utilize ultraviolet detection and none analyze both bupropion and hydroxybupropion in a single assay. Published methods for the detection of bupropion and its metabolites by tandem mass spectrometry are available, but none are chiral [26], [37]. In addition, there are no published assays for either chiral or achiral analysis of bupropion and its metabolites in urine. There is no information regarding the formation clearance of the hydroxybupropion enantiomers. Therefore, the purpose of this investigation was to develop and apply a stereoselective LC–MS–MS assay for the analysis, in human plasma and urine, of bupropion enantiomers and hydroxybupropion diastereomers in a single assay.

Section snippets

Chemicals and reagents

All standards were racemic and of >98% chemical purity. (±)-Hydroxybupropion was purchased from BD Gentest (San Jose, CA) and the internal standards (±)-bupropion-d9 and (±)-hydroxybupropion-d6 were purchased from Toronto Research Chemicals (Ontario, Canada). Oasis MCX 96-well solid phase extraction (SPE) plates were supplied by Waters (Milford, MA). Crude ß-glucuronidase (type HP-2), (±)-bupropion and all other reagents were ACS or reagent grade and purchased from Sigma Chemical Co. (St.

Mobile phase optimization

The LC method used in this paper was based initially on a previously published method for bupropion using an AGP column with UV detection [35]. When the ammonium acetate (pH 5.0) eluent from that method was applied to mass spectrometry, it resulted in poor ionization of bupropion and hydroxybupropion. By switching to ammonium formate (pH 5.7), the signal intensity was increased 100-fold. However, at pH 5.7, ammonium formate is not an ideal buffer and was not adequately buffering the sample,

Conclusion

This paper presents a novel LC–MS–MS method to analyze the enantiomers of both bupropion and hydroxybupropion in plasma and urine. The assay is linear in plasma from 0.5 to 200 ng/ml and 2.5 to 1000 ng/ml in each bupropion and hydroxybupropion enantiomer, respectively. The assay is linear in urine from 5 to 2000 ng/ml and 25 to 10,000 ng/ml in each bupropion and hydroxybupropion enantiomer, respectively. The predominant enantiomers in both urine and plasma were (R)-bupropion and (R,R

Acknowledgment

This study was supported by NIH R01DA14211 and K24DA00417.

References (39)

  • R.Z. Litten et al.

    J. Subst. Abuse Treat.

    (1999)
  • J.W. Jefferson et al.

    Clin. Ther.

    (2005)
  • Q.K. Fang et al.

    Tetrahedron: Asymmetry

    (2000)
  • J.S. Munro et al.

    J. Chromatogr. A

    (2001)
  • J. Hermansson et al.

    J. Chromatogr. A

    (1995)
  • H. Xu et al.

    Chirality

    (2007)
  • V. Borges et al.

    J. Chromatogr. B

    (2004)
  • K.K. Loboz et al.

    J. Chromatogr. B

    (2005)
  • J.R. Hughes et al.

    JAMA

    (1999)
  • A.J. Johnston et al.

    Drugs

    (2002)
  • A.A. Lai et al.

    J. Clin. Psychiatry

    (1983)
  • J. Posner et al.

    Eur. J. Clin. Pharmacol.

    (1984)
  • C.L. DeVane et al.

    J. Clin. Psychopharmacol.

    (1990)
  • P.H. Hsyu et al.

    J. Clin. Pharmacol.

    (1997)
  • S.C. Laizure et al.

    Clin. Pharmacol. Ther.

    (1985)
  • J. Posner et al.

    Eur. J. Clin. Pharmacol.

    (1985)
  • D.L. Musso et al.

    Chirality

    (1993)
  • M.I. Damaj et al.

    Mol. Pharmacol.

    (2004)
  • L.M. Hesse et al.

    Drug Metab. Dispos.

    (2000)
  • Cited by (0)

    View full text