Development of a quantitative LC–MS/MS analytical method coupled with turbulent flow chromatography for digoxin for the in vitro P-gp inhibition assay

https://doi.org/10.1016/j.jchromb.2007.04.030Get rights and content

Abstract

Caco-2 cells, the human colon carcinoma cells, are typically used for screening compounds for their permeability characteristics and P-glycoprotein (P-gp) interaction potential during discovery and development. The P-gp inhibition of test compounds is assessed by performing bi-directional permeability studies with digoxin, a well established P-gp substrate probe. Studies performed with digoxin alone as well as digoxin in presence of test compounds as putative inhibitors constitute the P-gp inhibition assay used to assess the potential liability of discovery compounds. Radiolabeled 3H-digoxin is commonly used in such studies followed by liquid scintillation counting. This manuscript describes the development of a sensitive, accurate, and reproducible LC–MS/MS method for analysis of digoxin and its internal standard digitoxin using an on-line extraction turbulent flow chromatography coupled to tandem mass spectrometric detection that is amendable to high throughput with use of 96-well plates. The standard curve for digoxin was linear between 10 nM and 5000 nM with regression coefficient (R2) of 0.99. The applicability and reliability of the analysis method was evaluated by successful demonstration of efflux ratio (permeability B to A over permeability A to B) greater than 10 for digoxin in Caco-2 cells. Additional evaluations were performed on 13 marketed compounds by conducting inhibition studies in Caco-2 cells using classical P-gp inhibitors (ketoconazole, cyclosporin, verapamil, quinidine, saquinavir etc.) and comparing the results to historical data with 3H-digoxin studies. Similarly, P-gp inhibition studies with LC–MS/MS analytical method for digoxin were also performed for 21 additional test compounds classified as negative, moderate, and potent P-gp inhibitors spanning multiple chemo types and results compared with the historical P-gp inhibition data from the 3H-digoxin studies. A very good correlation coefficient (R2) of 0.89 between the results from the two analytical methods affords an attractive LC–MS/MS analytical option for labs that need to conduct the P-gp inhibition assay without using radiolabeled compounds.

Introduction

P-glycoprotein (P-gp) is an ATP-dependent efflux transporter protein that is strategically located in several key tissues such as the small intestine, liver, kidney and blood–brain barrier. P-gp, by its ubiquitous expression coupled with the capacity to interact with a wide-spectrum of substrates, is known to play a prominent role in dictating the pharmacokinetics and pharmacodynamics of several drugs. It is widely recognized to be a major determinant of absorption, distribution and elimination of a wide array of marketed drugs [1].

In drug development, P-gp interaction potential of compounds is a factor in determining whether a test compound will be selected for further development. P-gp interaction could result from either the compound being a substrate or inhibitor for this important transporter. Caco-2 cell based digoxin transport inhibition assay is a well documented model to establish the drug candidate's potential to be a P-gp inhibitor [2], [3]. Radiolabeled 3H-Digoxin is typically used to perform the P-gp inhibition studies for test compounds. Current approaches to quantitate digoxin involve using radioactive 3H-digoxin in conjunction with liquid scintillation counting [4], [5], [6], [7], [8], [9], [10], [11]. There are several recurring costs associated with handling radioactive samples such as personnel trainings, spot checks, waste disposal etc. that can be a significant financial burden. The generation, storage and disposal of these radioactive samples present a formidable challenge to discovery organizations where this assay is being pushed upstream with more and more compounds being tested for their P-gp inhibition potential.

This manuscript presents efforts dedicated towards development of an LC–MS/MS detection method for digoxin using a turbulent flow chromatography technique to assay Caco-2 cell based bi-directional samples from the P-gp inhibition assay. This research was a follow-up from our earlier application of this analytical technique towards analysis of samples from Caco-2 cell based P-gp substrate assay [1].

Turbulent flow chromatography (TFC) interfaced with tandem mass spectrometry has been in use for the last decade. TFC provides an on-line extraction sample clean up that is automated and takes place in real time during the course of an injection thus eliminating manual sample preparation techniques such as liquid–liquid extraction, off-line solid phase extraction, and protein precipitation [12], [13], [14], [15], [16], [17], [18], [19], [20], [21]. Many recent publications utilizing TFC and the combined specificity of selected reaction monitoring through use of triple quadrupole mass spectrometry offers proof that this analytical approach is well accepted [22], [23], [24], [25], [26], [27], [28]. The development of a LC–MS/MS analytical method for digoxin for the P-gp inhibition assay has several other advantages: potential ease-of-transfer of the assay for higher throughput applications, only digoxin is being monitored thus the LC–MS/MS assay can be run continuously without optimizing for new compounds, there are no analytical issues associated with metabolism (i.e. degradation of digoxin) or tritium–water exchange. Thus, the specificity of MS/MS detection using tandem mass spectrometry affords an attractive alternative to eliminate the use of radioactive isotopes in this P-gp inhibition assay while providing high sensitivity.

Section snippets

Reagents and chemicals

Caco-2 cells were obtained from the American Type Culture Collection (Rockville, MD). Dulbecco's modified Eagle's medium, nonessential amino acids and Antibiotics were purchased from JHR Biosciences (Lenexa, KS). Fetal bovine serum was obtained from Hyclone Lab. Inc. (Logan, Utah). HTS-Transwell® inserts (surface area: 0.33 cm2 with a polycarbonate membrane (0.4 μm pore size) were purchased from Costar (Cambridge, MA). Hank's balanced salt solution (HBSS), N-2-hydroxyethylpiperazine-N

Standard curves, raw data, accuracy, and precision

Eight point standard curves in replicates of four were used to evaluate the overall accuracy and precision of the bioanalytical method. The standard curve ranged from 10 to 5000 nM for digoxin prepared in HBSS. This method development stage work was not incubated in Caco-2 cell lines and was only used to evaluate the linearity, sensitivity, and limits of quantitation for digoxin. The accuracy was defined as the percent difference from the nominal concentration. The mean was determined by

Discussion and conclusions

A bioanalytical method was evaluated in a pilot study of 34 different compounds on an Aria TX-2 TurboFlow® system from Thermo Scientific, using turbulent flow chromatography and tandem mass spectrometry. The use of TFC is very effective in de-salting samples with high levels of buffers and chelating agents. TFC as a sample preparation technique was automated and took place following sample injection; thereby it does not require any additional time to prepare samples off-line. The run time was

Reference (28)

  • J. Smalley et al.

    J. Chromatogr. B

    (2006)
  • P. Balimane et al.

    Eur. J. Pharm. Biopharm.

    (2004)
  • J. Keogh et al.

    Eur. J. Pharm. Sci.

    (2006)
  • M. Yao et al.

    J. Pharm. Biomed. Anal.

    (2003)
  • N. Yoshida et al.

    Food Chem. Toxicol.

    (2006)
  • A. Collett et al.

    Eur. J. Pharm. Sci.

    (2005)
  • A. Jedlicka et al.

    J. Pharm. Biomed. Anal.

    (2003)
  • J. Lin

    Adv. Drug Deliv. Rev.

    (2003)
  • M. Sababi et al.

    Eur. J. Pharm. Sci.

    (2001)
  • K. Kelly et al.

    J. Chromatogr. A

    (1995)
  • D. Zimmer et al.

    J. Chromatogr. A

    (1999)
  • M. Jemal et al.

    J. Pharm. Biomed. Anal.

    (2000)
  • A. Asperger et al.

    J. Chromatogr. A

    (2002)
  • L. Ynddal et al.

    J. Chromatogr. A

    (2003)
  • Cited by (0)

    View full text