Physico-chemical characterization of a novel group of dopamine D3/D2 receptor ligands, potential atypical antipsychotic agents

https://doi.org/10.1016/j.jpba.2008.06.021Get rights and content

Abstract

The fundamental physico-chemical properties such as ionization and lipophilicity of twelve alkyl–aryl-piperidine and aryl-piperazine derivatives have been determined. Compounds are members of a recently identified, new class of potent dopamine D3/D2 receptor ligands as potential atypical antipsychotic agents and were used in the development of a promising drug candidate (RGH-188) being present currently in clinical phase II investigations. The ionization constant (pKa) and the partition coefficient in octanol/water (log Poct) and cyclohexane/water systems (log Pch) were measured by validated analytical methods. Based on the highly precise physico-chemical data the structure–property relationships (SPR) were studied. The effect of the polar and apolar heteroatoms as well as polar and apolar surface areas on the partition in the two solvent systems was investigated by linear regression and multivariate linear regression analyses. Brain/blood concentration ratios (BB values) as a function of time were determined by HPLC analyses on plasma and brain homogenates of Wistar rats. A linear relationship has been found between Δlog P values (log Poct  log Pch) and experimental log BB values, verifying that physico-chemical data can predict pharmacokinetic behaviour.

Introduction

The physico-chemical properties, such as solubility, ionization (i.e., acidity, basicity), lipophilicity, permeability, H-bond donor/acceptor capacity, etc., have long been recognized as predictors of pharmacokinetic (ADME) parameters and have, therefore, been increasingly used in early stage of drug development [1], [2], [3], [4], [5].

The acid/base character sets the charge of a molecule in solution at a particular pH. It can be described by the dissociation constant (pKa). The knowledge of the charge state is necessary for the understanding of absorption, transport and receptor binding of drugs at the molecular level. Lipophilicity is another molecular property of immense importance in medicinal chemistry. The logarithm of octanol/water partition coefficient (log Poct) is the most extensively used parameter to quantitate lipophilicity. The log Poct value has been found to be a good predictor of the passive transport of drugs through the lipoidal membranes of the human body.

A further molecular property, the capability to penetrate through the blood–brain barrier (BBB) has fundamental importance in the drug design either the drug is intended to achieve high CNS or peripheral presence. The endothelial cells of the brain capillaries create the BBB, which is a specific physical barrier and a complex biochemical interface containing many physiological functions. Unfortunately, the in vivo experimental determination of brain/blood concentration ratio (Eq. (1)) is time-consuming, expensive and difficult. An additional problem is that in early screening of new chemical entities the synthesized material is not enough to carry out in vivo studies. In vitro biological models such as tissue culture monolayers may provide valuable additional data on transport processes in either direction beyond passive diffusion data; however, these methods are time-consuming, labor-intensive and have low throughput. Thus there is considerable interest in developing reliable, simple non-biological models for the prediction of blood–brain barrier permeation.BB=concentrationinbrainconcentrationinblood

Lots of effort has been made to replace the animal experiments with in vitro or in silico methods for prediction of BBB penetration [6]. The recently developed PAMPA (parallel artificial membrane permeability assay) method using porcine brain lipids [7] still needs standardization and validation. Evidently, it would be useful if the log BB values could be predicted by an easy-to-measure physico-chemical parameter, like log P. Young et al. [8] investigated 20 centrally acting H2 receptor antagonists and reported poor correlation between log BB and log Poct (r = 0.436). Better correlation was observed with cyclohexane/water partition coefficient (log Pch) (r = 0.732) and with the Δlog P parameter of Seiler (r = 0.831), calculated as a difference between log Poct and log Pch [9]. The Δlog P value is considered a measure of the hydrogen-bonding ability of a compound. The larger the Δlog P value, the more capable the molecule to form hydrogen-bonding. Consequently, the more hydrophilic the molecule will be, and can, therefore, better distribute toward blood than brain, since blood is known to be more water-like and less lipophilic than brain [10]. However, others have found conflicting results in case of hydroxypyridinones where Δlog P produced the weakest correlation and log Pch proved to be the better predictor of brain penetration [11].

In our previous work the complex physico-chemical profiling of sertraline, a centrally acting antidepressant drug was studied, allowing interpretation of its excellent pharmacokinetics. The almost complete absorption is governed by the high lipophilicity (i.e., log Poct = 4.30) but the good brain penetration (and the high log BB coefficient) can be explained in terms of Δlog P = 0 value, i.e., the lack of polar interactions [12].

In this paper we report a study on physico-chemical profiling of a recently synthesized set of compounds discovered for antipsychotic activity. The molecules containing aryl-piperidine or aryl-piperazine moiety (Fig. 1) act on dopamine D3/D2 receptors. There are several antipsychotic drugs in the therapy containing similar moieties, for example, aripiprazole, ziprasidone and risperidone, which display 5-TH2A > D2 antagonist activity. The hypothesis of this development was that (i) a moderate degree of D2 antagonism is a prerequisite to antipsychotic activity (ii) D3 antagonism may carry favourable effects such as cognitive enhancement and lack of catalepsy and (iii) in order to achieve simultaneous behavioural manifestation of D2 and D3 receptor antagonism the antipsychotic drug candidate should be more potent on D3 than on D2 receptors [13]. Several compounds were evaluated in a 3D-QSAR study using CoMFA based on their affinity to D3 receptors. The synthesis, the receptor affinity and the pharmacokinetic properties of some representatives were reported earlier [14]. RGH-188 was found as the most potent dopamine D3/D2 receptor antagonist/partial agonist in the series. The compound fulfils the above requirements, displayed high in vitro affinity at cloned human D3 receptors (Ki = 0.085 nM) with about six-fold less affinity for D2 receptors. It potently inhibited apomorphine-induced climbing in mice (ED50 = 0.27 mg/kg). RGH-188 did not produce catalepsy in rats up to 100-fold dose of its ED50 value [13]. Owing to its good pharmacologic features (beneficial cognitive effects) and favourable side effect profile (low risk of EPS liability), RGH-188 is considered to be a promising novel antipsychotic drug candidate, developed in cooperation between Gedeon Richter Plc. (Budapest, Hungary) and Forest Laboratories, Inc. (New York, USA) and being present currently under clinical phase II trials [15].

The objective of the present study was to determine the most important physico-chemical properties of 112 compounds from the antipsychotic D3/D2 project resulting finally RGH-188, including the aqueous dissociation constant (pKa), the partition coefficient (log P) in both octanol/water and cyclohexane/water solvent systems using validated methods. Based on the obtained, highly precise experimental data our purpose was to reveal the structure–property relationships (SPR), and to predict the pharmacokinetic properties, and to investigate the validation of log BB vs. Δlog P correlation. Our further aim was to understand the main molecular factors, which determine the partition processes of this series of compounds.

Section snippets

Materials

Samples of 112 were synthesized in Gedeon Richter Plc. (Budapest, Hungary) according to methods described in Hungarian [16], [17], [18] and US [19] patents and published partly in Ref. [14]. The purity of the samples was above 95%. The distilled water was of pharmacopeial grade [20] and all other reagents of analytical grade were purchased from commercial suppliers. The Britton-Robinson buffer (acetic, phosphoric and boric acids, each at 0.04 M, treated with 0.2 M NaOH) was used as the aqueous

Ionization

The examined compounds (112) differ considerably in acid–base properties representing a variety of proton-binding sites in both the type and the strength. There are monovalent bases (1, 2, 4, 68, 12) with piperidine (1, 2), piperazine (68, 12) and aniline nitrogen atom(s) (4). Compounds 3, 9, 10 are bivalent bases having two proton-binding sites: piperidine N and quinoline N (3); piperazine N and pyridine N (9) or quinoline N atom (10). Compound 5 has weak acidic character due to the

Conclusions

In this study the ionization and the lipophilicity of compounds 112 as members of a novel class of potent dopamine D3/D2 receptor ligands were described in terms of thermodynamic physico-chemical constants (pKa, log Poct and log Pch).

Based on physico-chemical data, the following compound-specific and compartment-selective pharmacokinetic behavior could be predicted and were eventually verified by the overall log BB values:

  • (i)

    Absorption from the stomach can be expected only for the weak acid,

Acknowledgement

The authors wish to thank Dr László Molnár for the calculation of PSA.

References (32)

  • E.H. Kerns et al.

    Drug Discov. Today

    (2003)
  • E.H. Kerns

    J. Pharm. Sci.

    (2001)
  • C.A. Lipinski

    J. Pharmacol. Tox. Meth.

    (2000)
  • N.J. Abbott

    Drug Disc. Today: Technol.

    (2004)
  • L. Di et al.

    Eur. J. Med. Chem.

    (2003)
  • M.H. Habgood et al.

    Biochem. Pharmacol.

    (1999)
  • I. Gyertyán et al.

    Eur. Neuropsychopharmacol.

    (2006)
  • É. Ágai-Csongor et al.

    Bioorg. Med. Chem. Lett.

    (2007)
  • K. Takács-Novák et al.

    Int. J. Pharm.

    (1997)
  • K.Y. Tam et al.

    Anal. Chim. Acta

    (2001)
  • K. Takács-Novák et al.

    J. Pharm. Biomed. Anal.

    (1996)
  • K. Takács-Novák et al.

    J. Pharm. Biomed. Anal.

    (1994)
  • K. Takács-Novák et al.

    Int. J. Pharm.

    (1995)
  • A. Avdeef et al.

    Cell. Mol. Life Sci.

    (2002)
  • A. Avdeef

    Curr. Top. Med. Chem.

    (2001)
  • R.C. Young et al.

    J. Med. Chem.

    (1988)
  • Cited by (11)

    View all citing articles on Scopus
    View full text