Elsevier

Toxicology in Vitro

Volume 30, Issue 1, Part A, 25 December 2015, Pages 192-202
Toxicology in Vitro

Amiodarone biokinetics, the formation of its major oxidative metabolite and neurotoxicity after acute and repeated exposure of brain cell cultures

https://doi.org/10.1016/j.tiv.2015.01.012Get rights and content

Highlights

  • Amiodarone biokinetics was studied in 2D and 3D brain cell models.

  • After single treatment Amiodarone cell uptake was completed after 24 h.

  • After repeated exposure Amiodarone bioaccumulates, especially in the 3D cell model.

  • Bioaccumulation is paralleled to some alterations of neurotoxicity markers.

  • The two brain cell models are able to form the Amiodarone major oxidative metabolite.

Abstract

The difficulty in mimicking nervous system complexity and cell–cell interactions as well as the lack of kinetics information has limited the use of in vitro neurotoxicity data.

Here, we assessed the biokinetic profile as well as the neurotoxicity of Amiodarone after acute and repeated exposure in two advanced rodent brain cell culture models, consisting of both neurons and glial cells organized in 2 or 3 dimensions to mimic the brain histiotypic structure and function. A strategy was applied to evidence the abiotic processes possibly affecting Amiodarone in vitro bioavailability, showing its ability to adsorb to the plastic devices. At clinically relevant Amiodarone concentrations, known to induce neurotoxicity in some patients during therapeutic treatment, a complete uptake was observed in both models in 24 h, after single exposure. After repeated treatments, bioaccumulation was observed, especially in the 3D cell model, together with a greater alteration of neurotoxicity markers. After 14 days, Amiodarone major oxidative metabolite (mono-N-desethylamiodarone) was detected at limited levels, indicating the presence of active drug metabolism enzymes (i.e. cytochrome P450) in both models.

The assessment of biokinetics provides useful information on the relevance of in vitro toxicity data and should be considered in the design of an Integrated Testing Strategy aimed to identify specific neurotoxic alerts, and to improve the neurotoxicity assay predictivity for human acute and repeated exposure.

Introduction

The nervous system is one of the most complex organ systems in terms of both structure and function; in addition, the lack of regeneration after severe damage renders the nervous system particularly vulnerable to toxic insult (Gramowski et al., 2004). Neurotoxicity is indeed one of the toxicity endpoints generally assessed in the safety evaluation of many chemicals as requested by EU Regulations, such as EC Regulations 1907/2006 (REACH); 1107/2009 (pesticides) and 528/2012 (biocides). Last but not least, the increasing onset of neuronal disorders and neurodegenerative diseases, linked to the aging of the population, represents a clear demand for drugs active on the nervous system for which safety has to be assessed in the early phase of development.

The currently accepted neurotoxicity studies include in vivo tests, usually on rodents (Bal-Price et al., 2010a); in this kind of studies direct adverse effects on the nervous system are often difficult to distinguish from indirect effects, linked to hormonal and immunological stimuli. This makes the interpretation of the observed functional changes quite difficult and leads to the conclusion that in vivo toxicity tests are not always ideal for the detection of neurotoxic effects (Harry and Tiffany-Castiglioni, 2005).

There is nowadays a large consensus that the animal testing needs to be replaced by a combination of in silico and in vitro approaches, as evidenced by ethical and economic arguments, besides scientific ones (Bal-Price et al., 2010a). However, so far in vitro studies have been proposed to be used as complementary to animal testing (Harry et al., 1998, Bal-Price et al., 2010b), essentially to support the development of biomarkers as early indicators of adverse effects (e.g. initial biochemical alterations), or the identification of mechanism of action. Indeed, none of the in vitro available tests could be considered as a standing alone method, relying on for the evaluation of the neurotoxic hazard of a chemical. This is mainly related to the complex anatomical structure of the nervous system, its physiology, as well as cell–cell interaction among different cell types (neuronal and glial) (Weiss, 2011). However, promising models to be used as alternatives to in vivo neurotoxicity testing have been proposed (Bal-Price et al., 2010b), addressing the issues of complexity and cell–cell interactions, such as (i) the primary cortical (glia-containing) murine neuronal network culture, established on the surface of micro-electrode arrays (neurochips) in a two dimensional structure (2D mouse model) (Weiss, 2011) and (ii) the tri-dimensional re-aggregating rat brain cell cultures (3D rat model), containing different brain cell types (Honegger and Zurich, 2011). The 2D mouse model allows to directly monitor changes of 200 features of the electrical firing patterns by deriving quantitative parameters belonging to five general categories: burst structure, oscillatory aspects, synchronicity, connectivity and general activity (Gramowski et al., 2010, Johnstone et al., 2010, Weiss, 2011). The 3D rat model is composed of neurons, astrocytes, oligodendrocytes and microglial cells, allowing multiple cell–cell interactions, and the development of histotypic structures such as extracellular matrix, synapses and myelinated axons (Zurich et al., 2004). It has been extensively used for neurotoxicological investigations, and several structural and functional endpoints have proved to be useful specific markers of neurotoxicity, such as the activity of cell type-specific enzymes, the expression of selected genes, as well as astroglial and microglial reactivities (Monnet-Tschudi et al., 1995, Zurich et al., 2004, Zurich et al., 2013).

The long-term stability of both models makes them suitable for both acute and repeated exposure to chemicals or drugs, giving in addition the possibility to study different functional endpoints, allowing to distinguish between general cytotoxicity and possible specific chemical-induced neurotoxicity.

The combination of the 2D and 3D brain cell models with a blood brain barrier (BBB) in vitro model (Culot et al., 2013) was evaluated as an improved in vitro neurotoxicity testing strategy (Schultz et al., 2015), aimed to overcome the lack of the in vivo filtering processes associated with the BBB or lipid-aqueous partitioning, contributing to the artificial nature of the culture exposure (Hallier-Vanuxeem et al., 2009). However, another important step forward to increase the predictability of in vitro testing is the measurement of in vitro kinetics, including all the abiotic and biological processes such as in situ metabolic capability, determining the actual exposure of cells, which can greatly influence the toxicological outcome, especially after repeated exposure (Coecke et al., 2013). Despite its relevance, this aspect is most of the times neglected, and the nominal concentration is used to refer to concentration-effect relationship.

Here we describe the in vitro biokinetics of Amiodarone (2-n-butyl-3-[3,5-diiodo-4-diethylaminoethoxybenzoyl]-benzofuran; AMI), the formation of its major oxidative metabolite mono-N-desethylamiodarone (MDEA), and some neurotoxicity end-points induced after acute and 14 day-repeated treatments in the 2D and 3D brain models. AMI, a Class III antiarrhythmic medication, is a non-competitive inhibitor of alpha- and beta-adrenergic receptors, commonly used to treat patients with refractory ventricular tachycardia and paroxysmal atrial fibrillation. In some patients under therapeutic treatment, during which the drug reaches plasma concentrations in the range of 1.3–2 μM (Lafuente-Lafuente et al., 2009), AMI is known to induce a variety of side effects including neurotoxicity (Brief et al., 1987), characterised by a set of symptoms such as headache, dizziness, fatigue, tremor, peripheral sensorimotor neuropathy, proximal muscle weakness, ataxia, with few cases of parkinsonism (Ishida et al., 2010).

Section snippets

Chemicals and reagents

AMI was purchased from Sigma–Aldrich (Switzerland, Germany and Italy, catalogue No A8423-1g, lot #109K1307), MDEA (purity: 98.9%) was synthesized and kindly provided by Sanofi-Aventis (Germany, catalogue No RDS 8262). Dimethylsulfoxide (DMSO) (purity ⩾99.9%; Sigma–Aldrich catalogue No 472301). For AMI and MDEA quantification HPLC grade chemicals were obtained from commercially available sources. The Milli-Q water purification system (Millipore, Italy) was used to obtain deionised water.

Stability, solubility and cross-contamination among wells of the test compound

Cytotoxicity testing

As shown in Fig. 1, the cell death induced by AMI treatment in the 2D mouse model was characterised by a flat dose–response in the range of the four tested concentrations (0.0035–10 μM), with apparently no differences between single and repeated exposure (although the latter is likely underestimated due to the limit of measuring released LDH over time). After a single treatment, AMI was not cytotoxic at 0.0035 μM, whereas at 0.350 and 1.25 μM AMI about 10–20% cell death was measured (no

Discussion

In vitro testing assessing the toxicity of chemicals and drugs has shown so far a limited capacity of correlation with in vivo effects, especially for neurotoxicants, due to both the complex structure and functionality of the nervous system and the lack of kinetics information.

These two issues were tentatively addressed within the European Seventh Framework Program Project Predict-IV (FP7/2009–2013) in the frame of which the present work was conducted. Biokinetic parameters in relation to

Conflict of Interest

The authors declare that there are no conflicts of interest.

Transparency Document

.

Acknowledgments

This project was funded by the European Union’s 7th Framework Programme under grant agreement no. 202222, Predict-IV. We thank Hannelore Popa-Henning for the administrative work and organization and Denise Tavel for technical support.

References (41)

  • F. Monnet-Tschudi et al.

    Microglial responsiveness as a sensitive marker for trimethyltin (TMT) neurotoxicity

    Brain Res.

    (1995)
  • G. Pomponio et al.

    In vitro kinetics of amiodarone and its major metabolite in two human liver cell models after acute and repeated treatments

    Toxicol. In Vitro

    (2015)
  • B.N. Singh

    Antiarrhythmic actions of amiodarone: a profile of a paradoxical agent

    Am. J. Cardiol.

    (1996)
  • J.M. Trivier et al.

    High-performance liquid chromatographic assay for amiodarone N-deethylation in microsomes of rat liver

    J. Chromatogr.

    (1992)
  • S.H. Wilson et al.

    Markers for gene expression in cultured cells from the nervous system

    J. Biol. Chem.

    (1972)
  • W.J. Wingo et al.

    Decarboxylation of l-glutamic acid by brain

    J. Biol. Chem.

    (1950)
  • A. Zahno et al.

    The role of CYP3A4 in amiodarone-associated toxicity on HepG2 cells

    Biochem. Pharmacol.

    (2011)
  • M.G. Zurich et al.

    Evaluation of aggregating brain cell cultures for the detection of acute organ-specific toxicity

    Toxicol. In Vitro

    (2013)
  • P.C. Adams et al.

    Amiodarone and its desethyl metabolite: tissue distribution and morphologic changes during long-term therapy

    Circulation

    (1985)
  • J.F. Brief et al.

    Distribution of amiodarone and its metabolite, desethylamiodarone, in human tissues

    Can. J. Physiol. Pharmacol.

    (1987)
  • Cited by (21)

    • Integrating biokinetics and in vitro studies to evaluate developmental neurotoxicity induced by chlorpyrifos in human iPSC-derived neural stem cells undergoing differentiation towards neuronal and glial cells

      2020, Reproductive Toxicology
      Citation Excerpt :

      Indeed, possible variations of test item concentration over time can be related to chemical reactions or limited stability, adsorption to plastic, binding to proteins and lipids present in cell culture medium or intracellular accumulation, especially in repeated-dosing studies [17–19]. These alterations may change the actual bioavailability of a compound, affecting its final adverse outcome [20]: therefore the assessment of in vitro kinetics is a pivotal element, especially for compounds that undergo bioactivation [18,19]. A step-wise procedure to characterise in vitro biokinetics was established within the European project PREDICT-IV (FP7-HEALTH program) [19,21–24], and has been recently endorsed in the OECD Guidance Document on Good In Vitro Method Practices (GIVIMP) [25].

    • Amiodarone does not affect brain injury in a rat model of transient forebrain ischemia

      2019, Medicina Intensiva
      Citation Excerpt :

      However, it is not clear whether the toxicity is a direct effect of amiodarone on neurons or a result of altered lysosomal function.15 The risk of neurotoxic effects may increase with the duration of amiodarone treatment.5,6 Exposure limited to an early recovery period from ischemia might not deteriorate ischemic neuronal injury as much as expected.

    • Biokinetics in repeated-dosing in vitro drug toxicity studies

      2015, Toxicology in Vitro
      Citation Excerpt :

      A dynamic equilibrium between the concentration in cell and medium after repeated dosing was generally reached around 7 days after the initial dose. Pomponio et al. (2015a,b) measured amiodarone concentrations and its major metabolite MDEA in primary human hepatocytes, HepaRG, 2D mouse and 3D aggregating rat brain cell cultures. Uptake of amiodarone into hepatocytes was quicker than in brain cultures, which can be explained by the metabolism of the drug by the hepatocytes simultaneously occurring during the uptake phase of the drug into cells.

    View all citing articles on Scopus
    1

    These authors equally contributed to the work.

    View full text