The fast-off hypothesis revisited: A functional kinetic study of antipsychotic antagonism of the dopamine D2 receptor

https://doi.org/10.1016/j.euroneuro.2016.01.001Get rights and content

Abstract

Newer, “atypical” antipsychotics carry a lower risk of motor side-effects than older, “typical” compounds. It has been proposed that a ~100-fold faster dissociation from the dopamine D2 receptor (D2R) distinguishes atypical from typical antipsychotics. Furthermore, differing antipsychotic D2R affinities have been suggested to reflect differences in dissociation rate constants (koff), while association rate constants (kon) were assumed to be similar. However, it was recently demonstrated that lipophilic accumulation of ligand in the cell interior and/or membrane can cause underestimation of koff, and as high-affinity D2R antagonists are frequently lipophilic, this may have been a confounding factor in previous studies. In the present work, a functional electrophysiology assay was used to measure the recovery of dopamine-mediated D2R responsivity from antipsychotic antagonism, using elevated concentrations of dopamine to prevent the potential bias of re-binding of lipophilic ligands. The variability of antipsychotic kon was also reexamined, capitalizing on the temporal resolution of the assay. kon was estimated from the experimental recordings using a simple mathematical model assumed to describe the binding process. The time course of recovery from haloperidol (typical antipsychotic) was only 6.4- to 2.5-fold slower than that of the atypical antipsychotics, amisulpride, clozapine, and quetiapine, while antipsychotic kons were found to vary more widely than previously suggested. Finally, affinities calculated using our kon and koff estimates correlated well with functional potency and with affinities reported from radioligand binding studies. In light of these findings, it appears unlikely that typical and atypical antipsychotics are primarily distinguished by their D2R binding kinetics.

Introduction

The dopamine D2 receptor (D2R), a G protein-coupled receptor which signals via inhibitory Gi/o proteins, is crucially involved in psychotic disorders such as schizophrenia. Traditionally, antipsychotics have been divided into typical (e.g., haloperidol and chlorpromazine) and atypical (e.g., clozapine, quetiapine, and amisulpride) compounds based on their propensity to generate motor side-effects (extrapyramidal symptoms; EPS). These often severe adverse effects diminish the willingness of patients to comply with medication, and thus constitute a major limitation to the clinical utility of antipsychotics.

The lower liability of the atypical antipsychotics to give rise to these side-effects has been suggested to be a consequence of their faster dissociation rates (koffs) from D2R, which would produce rapidly reversible antagonism, preserving the physiological dynamics of D2R signaling (Kapur and Seeman, 2001). The idea that a fast koff is associated with a favorable side-effect profile has prompted several pharmaceutical companies to develop new fast-dissociating D2R antagonists (Dyhring et al., 2010, Koprich et al., 2013, Langlois et al., 2012, Pompeu et al., 2015). Moreover, radioligand dissociation experiments indicated that the wide variability (>100-fold) in D2R affinities of clinically used antipsychotics could largely be accounted for by differences in koff and since affinity, expressed as the dissociation constant Kd, is determined by the ratio of dissociation and association rate constants (Kd=koff/kon), differences in kon should be negligible (Kapur and Seeman, 2000). Hence, by extension of the “fast-off” hypothesis cited above, low-affinity D2R antagonists were postulated to be atypical antipsychotics. However, kon was not measured directly, but only estimated from the ratio koff/Kd (Kapur and Seeman, 2000).

We recently investigated the relative rates of D2R response recovery from antagonism by a range of antipsychotics using a Xenopus oocyte electrophysiology assay (Sahlholm et al., 2014). This assay is based on the activation of G protein-coupled inwardly rectifying potassium channels (GIRK) by Gβγ subunits liberated from Gαi/o proteins activated by the D2R. The fraction of open channels serves as a functional readout of receptor occupancy by dopamine (DA) with second-scale temporal resolution. We reported that the differences in rates of recovery from inhibition by the typical antipsychotic, chlorpromazine, and the atypical compounds, amisulpride, clozapine, and quetiapine, were on the order of 2-fold rather than 100-fold as suggested by the proponents of the fast-off hypothesis (Kapur and Seeman, 2001). We also observed that ligands with high calculated lipophilicity and low water solubility exhibited a component of long-lasting antagonism which could not readily be overcome by replacing the extracellular buffer. This is in agreement with recent reports that lipophilic membrane accumulation and subsequent re-binding of ligand to its receptor can lead to falsely low estimates of ligand koff (Packeu et al., 2010). Lipophilic ligands such as haloperidol have been shown to cross- or accumulate in-cell membranes, whereas hydrophilic ligands such as sulpiride do not (Packeu et al., 2010, Rayport and Sulzer, 1995). Thus, by diffusing from within the cell interior or laterally in the membrane, the effective concentration of antagonist might remain high at D2R receptors on the cell surface even after it has been washed out from the extracellular buffer.

Given the impact of the “fast-off” hypothesis on ongoing efforts to create improved antipsychotics, we explored the long-lasting D2R antagonism afforded by some highly lipophilic compounds such as haloperidol. Moreover, we reexamined the variability of kon among different antipsychotics, capitalizing on the high temporal resolution of the GIRK assay. To estimate kon, a mathematical expression was derived based on a simple three-state kinetic scheme, assuming binding of DA and antagonist to the same receptor site. Finally, the usefulness of our kinetic measurements was tested by calculating kinetic Kd values for the antipsychotics in question, and comparing these Kd values with inhibition constants (Kis) obtained both from equilibrium radioligand binding studies reported in the literature and from cumulative inhibition experiments using the GIRK assay.

Human GIRK1 (Kir3.1) and GIRK4 (Kir3.4) cDNA (provided by Dr. Terence Hebert, University of Montreal, Canada) and RGS4 (from the Missouri cDNA Resource Center; www.cdna.org) were in pCDNA3 (Invitrogen). cDNA encoding the human dopamine D2S and D2L receptors were in pXOOM (a gift from Dr. Søren-Peter Olesen, University of Copenhagen, Denmark). For in vitro transcription, plasmids were linearized with the appropriate restriction enzymes (GIRK 1/4, NotI; RGS4 and D2S/D2L, XhoI) and transcribed in vitro using the T7 mMessage mMachine kit (Ambion, Austin, TX). cRNA concentration and purity were determined using a spectrophotometer.

Oocytes were surgically isolated from female African clawed toads, Xenopus laevis, or purchased from EcoCyte Bioscience (Castrop-Rauxel, Germany), and injected with cRNA as previously described (Sahlholm et al., 2011). The surgical procedures have been approved by the Swedish National Board for Laboratory Animals. 1 ng of each GIRK1/4 subunit cRNA, 40 ng of RGS4 cRNA, and 0.2 ng of dopamine D2S or D2L receptor cRNA were injected per oocyte.

Clozapine, DA, and paliperidone were purchased from Sigma-Aldrich (St. Louis, MO), asenapine, chlorpromazine, haloperidol, N-desmethylclozapine, and olanzapine were from Abcam chemicals (Cambridge, UK), amisulpride and quetiapine were from Axon MedChem BV (Groeningen, The Netherlands), and remoxipride was a gift from Astra (Södertälje, Sweden). JNJ-37822681 was custom synthesized by Axon MedChem BV. Amisulpride, asenapine, chlorpromazine, DA, JNJ-37822681, and remoxipride were dissolved in distilled water, whereas the other ligands were dissolved in DMSO. Ligands were diluted in the recording solution to obtain the desired concentrations. The maximum final concentration of DMSO used in any experiment did not exceed 0.3% v/v.

The electrophysiological experiments were performed at room temperature (20–22 °C), 5–7 days after cRNA injection using a two-electrode voltage-clamp setup (CA-1 amplifier, Dagan, Minneapolis, MN) as previously described (Sahlholm et al., 2011). Data were acquired at 134 Hz using Molecular Devices software (pClamp 8.2). A high-potassium solution (in mM; 64 NaCl, 25 KCl, 0.8 MgCl2, 0.4 CaCl2, 15 HEPES, 1 ascorbic acid, adjusted to pH 7.4), giving a K+ reversal potential of about −40 mV, was used for GIRK current recording. Ascorbic acid was present in order to prevent the oxidation of DA. Single −80 mV pulses were applied to study GIRK current responses to D2R activity. Ligands were added to the 20 µl recording chamber by superfusion at 1.5 ml/min using a pressure-driven, computer-controlled perfusion system (SmartSquirt; AutoMate Scientific, Berkeley, CA).

During the response recovery protocols, 1 µM DA was applied to provide a baseline response, after which the antagonist of interest was applied in the continued presence of DA. To obtain response recovery data, the antipsychotic ligand was applied during 125 s, after which the antipsychotic was washed out while DA was increased to 100 µM or 1 mM. For kon estimation, the baseline response was evoked by 100 nM DA, and variable concentrations of antagonist were applied during 125 s in order to record the rate of response inhibition. For antagonist concentration-response data, following an initial application of 100 nM or 1 µM DA, three to five increasing concentrations of the antipsychotic ligand were applied consecutively, in the continued presence of DA, at 50 s intervals. For asenapine, haloperidol, and paliperidone, which displayed slow inhibition kinetics, 100 s intervals were used instead. While even longer antagonist applications might have been ideal for some of these antagonists, we limited ourselves to 100 s applications in order to avoid rundown of GIRK currents due to changes in intracellular Na+ concentrations, which have been shown to occur during prolonged recordings (Vorobiov et al., 1998). For each oocyte, the current amplitude at the end of each antagonist application interval was normalized to the control response to DA alone in the same oocyte. For DA concentration-response data, 4–5 increasing concentrations of DA were applied at 25 s intervals, ending with a response-saturating concentration (100 µM) of DA. For each oocyte, the DA-evoked current response to each concentration was normalized to the response evoked by 100 µM DA obtained in the same oocyte. In all electrophysiology protocols, the DA-evoked current response was determined by subtracting the basal (agonist-independent) current from the experimental record.

Variable slope sigmoidal concentration-response curves were fitted to the concentration-response data using GraphPad (Prism Software). The following equation was used for fitting:Y=11+10n(log10IC50X)where Y is the response as a fraction of 1, X is the decadic logarithm of ligand concentration and n is the Hill slope. When fitting the data for quetiapine and N-desmethylclozapine, n was constrained to −1. Clampfit (Molecular Devices) software was used for analyzing current traces, including determining rates of response recovery and inhibition. Water solubility (cLogS) and lipid/water distribution (cLogD) coefficients at pH 7.4 were calculated from antagonist structures using online prediction tools; ACD/I-Lab (Advanced Chemistry Development Inc., Toronto, Canada; https://ilab.acdlabs.com).

Assume that the binding of DA and antagonist is described by the following simplified schemexα×[D]βykon×[A]koffzwhere x represents the DA-bound state, y the unoccupied state and z the antagonist-bound state of the receptor; where α, β, kon and koff denote rate constants, and [D] and [A] represent DA and antagonist concentrations, respectively. Assume further that the transition between x and y is much faster than that between y and z, yielding the approximation x=y×α×[D]/β. Since x+y+z=1, we gety=1zy×α×[D]/βyieldingy=1z1+α×[D]/β=R0×(1z)where R0 is β/(α×[D]+β), i.e., the fraction of unoccupied receptors at the moment immediately preceding antagonist application (at t=0).

The differential equation describing the rightmost transition of Scheme 1 isdzdt=kon×[A]×ykoff×z.

Substituting y with Eq. (3) yieldsdzdt=kon×[A]×R0×(1z)koff×z

This differential equation has a solution with a single exponential where the exponent is kobs×t. Solving for kobs we getkobs=kon×[A]×R0+koffwhich is the inverse time constant of the antagonist effect.

In the present work, kobs and koff were measured directly. kon was obtained by the linear slope of the dependence of kobs on the antagonist concentration, [A], i.e.,kon=ΔkobsΔ[A]×R0

Finally, the dissociation constant Kd was obtained as Kd=koff/kon

Section snippets

Results

The EC50 for DA-induced GIRK activation under our experimental conditions was 17 nM (Fig. 1A).The antagonist concentrations used in the response recovery experiments (see below) were chosen based on concentration-response relationships obtained from cumulative inhibition experiments (Fig. 1B; Table 1), to achieve near-maximal inhibition. In the case of chlorpromazine, clozapine, N-desmethylclozapine, and quetiapine this was not possible due to channel blocking effects of these drugs at high µM

Discussion

In the present work, a relatively rapid recovery of D2R responsivity from antagonism by the typical antipsychotic, haloperidol, is described. While about 6.4- and 3.9-fold slower than the rates of recovery from clozapine and quetiapine, this is still much faster than the ~100-fold slower dissociation rate reported earlier for haloperidol relative to the former two antipsychotics (Kapur and Seeman, 2000). Additionally, as has been previously pointed out (Meltzer et al., 2003), Kapur and Seeman

This work was supported by research grants from the Swedish Research Council (04x-715 to KF, 15083 to PÅ, and 21785-01-4 to JN), Karolinska Institutet funds, Magn. Bergvalls stiftelse, Lars Hiertas Minne, and Åhlénstiftelsen (to KS). KS is a recipient of postdoctoral fellowships from the Swedish Society for Medical Research and the Swedish Brain Foundation. The funding sources had no further role in study design, in the collection, analysis, and interpretation of data, in manuscript writing, or

Acknowledgments

The authors thank AstraZeneca (Södertälje, Sweden) for the gift of remoxipride. Sofia Frisk, MD, Karolinska University Hospital, Stockholm, Sweden, is acknowledged for technical assistance during initial GIRK assay experiments.

References (28)

  • Y. Cheng et al.

    Relationship between the inhibition constant (K1) and the concentration of inhibitor which causes 50 per cent inhibition (I50) of an enzymatic reaction

    Biochem. Pharmacol.

    (1973)
  • C.P. Ford et al.

    The time course of dopamine transmission in the ventral tegmental area

    J. Neurosci.

    (2009)
  • S. Kapur et al.

    Antipsychotic agents differ in how fast they come off the dopamine D2 receptors. Implications for atypical antipsychotic action

    J. Psychiatry Neurosci.

    (2000)
  • S. Kapur et al.

    Does fast dissociation from the dopamine d(2) receptor explain the action of atypical antipsychotics?: a new hypothesis

    Am. J. Psychiatry

    (2001)
  • Cited by (35)

    • Clozapine's multiple cellular mechanisms: What do we know after more than fifty years? A systematic review and critical assessment of translational mechanisms relevant for innovative strategies in treatment-resistant schizophrenia

      2022, Pharmacology and Therapeutics
      Citation Excerpt :

      In fact, these authors have shown that the Koff of typical antipsychotics may be often biased and underestimated due to the lipophilic nature of typical compounds, allowing for sequestration of the antipsychotic in the cell membrane or cell interior, which then facilitates subsequent rebinding to D2Rs (Sahlholm et al., 2016). In this perspective, Sahlholm et al reported that clozapine’s dissociation is only 6-fold faster in comparison to typical antipsychotics, such as haloperidol, instead of 100-fold faster as suggested in previous receptor binding kinetic assays (Sahlholm et al., 2016). Therefore, factors other than D2R kinetic binding, probably involving a specific moiety of the compound, may participate in the specific clozapine dopaminergic action profile.

    • Continuous versus extended antipsychotic dosing in schizophrenia: Less is more

      2021, Behavioural Brain Research
      Citation Excerpt :

      Below we lay out the evidence that extended dosing schedules producing regular but transient D2 occupancy above 65 % are as effective as conventional, continuous dosing schedules in preventing relapse. Atypical antipsychotic drugs generally dissociate more rapidly from D2 receptors (i.e., have a faster koff) than typicals do [60] (but see [61]), but atypicals are still efficacious in treating schizophrenia symptoms. This has led to the hypothesis that antipsychotic efficacy could be achieved even with extended antipsychotic dosing.

    View all citing articles on Scopus
    View full text