Article pubs.acs.org/ac
HPLC/ICP-MS in Combination with “Reverse” Online Isotope Dilution in Drug Metabolism Studies Björn Meermann,† Anne Hulstaert,† Aline Laenen,‡ Cis Van Looveren,‡ Maarten Vliegen,‡ Filip Cuyckens,‡ and Frank Vanhaecke*,† †
Department of Analytical Chemistry, Ghent University, Krijgslaan 281-S12, 9000 Ghent, Belgium Drug Safety Sciences, Janssen R&D, Turnhoutseweg 30, 2340 Beerse, Belgium
‡
ABSTRACT: During the development of a new drug compound, its metabolism needs to be unraveled. For quantification of the metabolites formed, the drug under investigation is traditionally synthesized with a radiolabel (14C or 3H) and the metabolites present in different matrixes (blood, urine, feces) upon drug administration are determined by means of high-performance liquid chromatography (HPLC) coupled to radiodetection. This approach allows for quantification of the metabolites formed and enables a straightforward distinction between exogenous (i.e., drug-related) and endogenous species (as only the radiolabeled species are detected). However, in some cases, the use of a radiolabeled compound in human in vivo studies is not advisible, e.g., for drug compounds or their metabolites showing a long plasma or tissue half-life. In cases where the candidate drug molecule contains an element detectable by means of inductively coupled plasma mass spectrometry (ICP-MS), HPLC/ICP-MS is a promising alternative approach. However, the method lacks specificity when a distinction between drug-related species and endogenous compounds containing the same target element needs to be accomplished. As a result, we have developed an HPLC/ICP-MS-based method combined with “reverse” online isotope dilution (“reverse” online ID) for metabolite quantification. The methodology was evaluated by the analysis of feces samples from rats dosed with a 81Br-labeled drug compound. The method allows for both (i) valid quantification of the drug metabolites and (ii) distinction among endogenous, exogenous, and “mixed” species, based on their isotopic “fingerprint”. A good repeatability (relative standard deviation of 4.2%) and limit of detection (0.35 mg of drug compound L−1 of feces extract), of the same order of magnitude as those observed for “normal” online ID HPLC/ICP-MS and HPLC/radiodetection, were achieved.
I
However, combining RP-HPLC relying on gradient elution with ICP-MS detection is a challenging task, due to the fact that signal or sensitivity drift typically occurs within the chromatogram recorded. As a result, drift-compensating strategies, such as the use of a cooled spray chamber5 or the addition of a countergradient,6 need to be deployed here; further approaches to overcome this drift are discussed in detail elsewhere.7,8 One promising approach enabling drift compensation and direct quantification of elements possessing at least two stable isotopes is based on online isotope dilution (online ID).9 We have recently reported on the application of RP-HPLC/ICPMS combined with online ID for the successful speciation analysis of the metabolites of a bromine-containing antituberculosis drug in the context of a human in vivo study.10 In this approach, all species that contain the “heteroelement” under investigation (e.g., bromine) are detected. Due to the fact
n vitro and in vivo metabolism studies form a crucial part within the development of new drug components. Analytical methods that enable identification and quantification of the metabolites are indispensable. Analytical techniques meeting this demand comprise a separation systemtypically (reversed-phase, RP) high-performance liquid chromatography (HPLC)coupled to electrospray ionization mass spectrometry (ESI-MS) to obtain structural information and to radiodetection to enable quantification.1 Applying radiodetection for metabolite quantification necessitates radioactive labeling of the drug under investigation (3H or 14C label). Although the radioactive doses involved are low, problems can occur in the case of drugs or metabolites showing a long tissue/ plasma half-life. As a result, for drugs containing an element detectable by means of inductively coupled plasma mass spectrometry (ICP-MS), some research groups demonstrated the coupling of (RP) HPLC to ICP-MS to be an alternative, promising approach,2−4 making a radioactive label dispensable while providing similar or higher sensitivity. © 2012 American Chemical Society
Received: November 29, 2011 Accepted: February 2, 2012 Published: February 2, 2012 2395
dx.doi.org/10.1021/ac203165p | Anal. Chem. 2012, 84, 2395−2401
Analytical Chemistry
Article
trile (LC−MS Chromasolv) and dimethyl sulfoxide (DMSO; puriss p.a.) were obtained from Riedel-De Haen (Seelze, Germany). For the preparation of “spike A”, used for the analysis of the samples containing bromine of natural isotopic composition (natBr) only, 99.62% enriched 81Br (Na81Br) from CK Gas Products Ltd. (Hampshire, Great Britain) and sodium bromide of natural isotopic composition (pure, Merck, Darmstadt, Germany) were used. For the preparation of “spike B”, used for the analysis of the samples containing drug-related species enriched in 81Br, a certified ICP-MS bromine elemental standard (1006 ± 3 μg mL−1 bromine) from Inorganic Ventures Inc. (Lakewood, NJ) was used. The water used was purified by means of a Direct-Q3 system (Millipore, Molsheim, France). Oxygen (Alphagaz 2) was obtained from Air Liquide (Liège, Belgium). Buffer and Standard Solutions. Gradient elution during RP-HPLC separation was accomplished using two different solvents: Solvent A consisted of 0.1 mol L−1 ammonium acetate buffer (pH ≈ 7.1). Solvent B consisted of a mixture of 10% 1 mol L−1 ammonium acetate buffer (pH ≈ 7.1), 45% methanol, and 45% acetonitrile. The gradient applied is listed in Table 1. The 81Br spike solution (spike A) for the normal online isotope dilution approach (samples containing natBr only) was prepared using 81Br-enriched sodium bromide and sodium bromide containing natBr. Two aqueous stock solutions (natBr versus 81Br), each containing 1 g of Br L−1 were prepared. A 1.61 mL volume of 81Br stock solution was mixed with 0.39 mL of natBr stock solution and the mixture thus obtained made up with water to 1 L to end up at a final concentration of 2 mg of Br L−1 and a 81Br/79Br isotope ratio of ∼9/1. The natBr spike solution (spike B) for the reverse online isotope dilution approach (samples containing 81Br-labeled drug-related compounds) was prepared using a natBr-containing elemental standard solution (1006 ± 3 μg mL−1 bromine). From this stock solution, an aqueous solution containing 2 mg of Br L−1 with the natural Br isotope ratio 81Br/79Br of ∼1/1 was prepared by dilution. Rat in Vivo Study. Formulations. Appropriate amounts of 14C-labeled TMC207 (containing natBr) (see Figure 1) with a specific activity of 1.93 GBq mmol−1 (or 3.48 MBq mg−1 or 94 μCi mg−1 with respect to the molar mass of 555.51 g mol−1) and [natBr]TMC207 were mixed, and the mixture thus obtained was diluted with methanol to obtain a specific activity of approximately 247 kBq mg−1. The radiochemical purity of 14Clabeled TMC207 was ≥95%. Thereafter, the solvent was evaporated to dryness, and the residue was used to prepare “formulation A”. “Formulation B” was prepared in an identical way, except for the replacement of [natBr]TMC207 by TMC207 enriched in 81Br (see Figure 1B). The synthesis route for TMC207 can be found elsewhere.15 In the case of formulation B, bromine enriched in 81 Br was used. Formulations A and B were prepared one day prior to dose administration. An appropriate amount of aqueous 20% (hydroxypropyl)-β-cyclodextrin (HP-β-CD) (equivalent to half of the final formulation volume) was acidified by the addition of 1 M HCl. The total amount of test substance (A, radiolabeled and natBr; B, radiolabeled and 81Br) in the evaporation residue was dissolved in the acidified HP-β-CD solution to obtain a final concentration of 2.00 ± 0.10 mg of TMC207 mL−1. Afterward, the pH was adjusted to 2.0 ± 0.1 by slow addition of 1 M NaOH.
that the study was carried out within the scope of a clinical trial, appropriate blank samples were available and a distinction between endogenous and exogenous (i.e., drug-related) species was possible. If the demand of appropriate blank samples is not fulfilled, misinterpretation of the results followed by nonvalid quantification will occur. To ensure a valid quantification notwithstanding the simultaneous presence of endogenous compounds containing the same heteroelement, a modified online ID approach was investigated. Within this approach, a version of the drug containing the heteroelement under investigation (e.g., bromine) in a non-natural isotopic composition was synthesized, while the spike solution admixed with the column effluent for ID purposes then contains the heteroelement in its natural isotopic composition instead. Of course, also the synthesis of a 81Br-labeled drug analogue is expensive and time-consuming; however, the costs and the effort of synthesis are similar to those of the radio-labeling approach, while the “reverse” online ID approach offers further advantages. In cases wherein no appropriate blank samples are available or samples are contaminated (e.g., with halogenated flame retardants,11,12 which are increasingly emitted into the environment, or other halogenated anthropogenic substances such as pesticides13,14), reverse online ID allows a distinction between endogenous (blank) and exogenous (drug-related compounds) to be made on the basis of the isotope ratio displayed by the target element. Furthermore, even in the case of coelution or adduct formation between endogenous and exogenous species, a distinction can be made. Besides this, a quantification of all different types of species (endogenous, exogenous, and intermediate) is possible by determining the respective isotope ratio and adapting the spike solution. In the present work, a bromine-containing antituberculosis drug (TMC207) was labeled with a 81Br label and administered to rats within an in vivo study. Feces samples were collected and analyzed by means of HPLC/ICP-MS with the reverse online ID method developed. Furthermore, the results were compared to those obtained with the “normal” online ID HPLC/ICP-MS approach for samples from a rat control group that were given TMC207 containing Br of natural isotopic composition ([natBr]TMC207). For further validation purposes both formulations contain a 14C label as well (Figure 1) enabling HPLC/radiodetection measurements.
Figure 1. Structure of the bromine-containing antituberculosis drug TMC207 with (A) natBr and 14C label and (B) 81Br and 14C label. The position of the 14C label is marked with an asterisk.
■
EXPERIMENTAL SECTION
Chemicals. Ammonium acetate (99.99+%, total metals
