Xiao Zhu (1), David B Finlay (2), Michelle Glass (2), Stephen B Duffull (1)
(1) Otago Pharmacometrics Group, School of Pharmacy, University of Otago, Dunedin, New Zealand; (2) Centre for Brain Research and Department of Pharmacology and Clinical Pharmacology, Faculty of Medical and Health Sciences, University of Auckland, Auckland, New Zealand
Objectives:
Cannabinoid type 1 (CB1) receptor is G-protein coupled receptor (GPCR) present at high levels throughout the CNS [1]. CB1 is an attractive therapeutic target for numerous central nervous system diseases, including neurodegenerative disease, multiple sclerosis and pain. However, despite the enormous therapeutic potentials, clinical application of CB1 ligands has been hampered owing to their adverse on-target effects. Recently, interests have risen in the ability of ligands to differentially regulate multiple signalling pathways when coupled to a single receptor, termed biased agonism [2]. One emerging therapeutic strategy is to improve the selectivity and safety of drugs through the development of ligands that are biased towards selected pathways. Therefore, it is crucial to have a comprehensive understanding of ligand-biased signalling at CB1.
The time course of the stimulus response system often means the bioassay of the pathway of interest does not achieve equilibrium. This is compounded by receptor internalisation which under positive ligand activation occurs over the same time course as other pathway measurements. Currently these experiments are analysed by using single time-point choice to represent a snapshot of the process at a time where the investigator believes an equilibrium has occurred and then assesses each pathway separately. A full kinetic model is required in order to quantitatively assess the time-dependent modulation of the receptor by ligands and provide novel insights into the complex interplay among ligands, receptors and pathways.
The aims of current study was to develop a mechanism-based model that quantitatively describes the time course data from internalisation, phospho-ERK (pERK), and cyclic AMP (cAMP) pathways coupled to CB1 receptor.
Methods:
Data
The internalisation was determined utilizing a live cell antibody feeding technique[3]. The terminal sampling was performed at 0, 2, 4, 8, 15, 30, and 60 min. Quantification of pERK was performed using the AlphaScreen SureFire kit (Perkin Elmer) [4]. The terminal sampling was performed at 1, 3, 4, 5, 6, 8, 12, 20, 40 and 60 min. cAMP was measured using a kinetic BRET assay (CAMYEL) [3]. Forskolin (FSK) was added to stimulate the synthesis of cAMP. The real time sampling was performed every 0.4-0.5 min for 20 min.
All the assays were performed under multiple concentration levels of six CB1 ligands.
Model
The stimulus response model incorporated a transducer function (f, rectangular hyperbola in this study) to convert stimulus (S, resulting from receptor occupancy) into response, which allowed the response to represent a potential cascade of cellular or tissue signalling [5].
A mechanism-based stimulus response model was developed to describe the time course of three pathways sequentially using a PPP&D modelling framework:
1) Internalisation was described by a target-mediated drug disposition model with a quasi-steady state assumption [6].
2) pERK was described by an indirect response model [7]. The ligand effect was added via stimulus response model with stimulation on the synthesis rate.
3) cAMP was described by an indirect response model with capacity limited stimulation by FSK ([Emax*FSK]/[EC50+FSK]) on the synthesis rate of cAMP [7]. The ligand effect was added via a stimulus response model with inhibition on the synthesis rate.
Due to the intensive sampling in cAMP measurement, an AR(1) model was used to account for correlation in the residual errors[8].
The model was implemented in the ADVAN13 subroutine in NONMEM 7.3 [9]. The estimation method was FOCE-I.
Results:
The developed model adequately described the observed data. All model parameters were precisely estimated (<50% RSE) and within physiological ranges or agreed with values reported in the literature. The internalisation rate constant (from 0.028 to 1.05 min-1) was generally much faster than the receptor degradation constant (0.0015min-1, 15% RSE). For the pERK pathway, the estimated system maximal stimulation was 68.8 fold over baseline (11% RSE). The estimated duration of stimulation was 3.68min (2% RSE), which was consistent with observed peak time (from 3 to 5 min). For the cAMP pathway, the estimated system maximal inhibition was 0.718 fold over baseline (3% RSE).
Conclusions:
The joint model provided a reasonable description of the dynamic change of CB1 receptor, leading to a better understanding of CB1 system. In so doing, it could facilitate the discovery of novel therapeutics with improved selectivity and safety.
References:
[1] Glass M, Dragunow M, Faull RL. Cannabinoid receptors in the human brain: a detailed anatomical and quantitative autoradiographic study in the fetal, neonatal and adult human brain. Neuroscience. 1997; 77: 299–318.
[2] Urban JD, Clarke WP, von Zastrow M, Nichols DE, Kobilka B, Weinstein H, et al. Functional selectivity and classical concepts of quantitative pharmacology. J Pharmacol Exp Ther. 2007; 320: 1-13.
[3] Cawston EE, Redmond WJ, Breen CM, Grimsey NL, Connor M, Glass M. Real-time characterization of cannabinoid receptor 1 (CB1) allosteric modulators reveals novel mechanism of action. Br J Pharmacol. 2013; 170: 893–907.
[4] Finlay DB, Cawston EE, Grimsey NL, Hunter MR, Korde A, Vemuri VK, Makriyannis A, Glass M. Gαs signalling of the CB1 receptor and the influence of receptor number. Br J Pharmacol. 2017; 174:2545-2562.
[5] Visser SAG, Smulders C, Reijers BPR, van der Graaf PH, Peletier LA, Danhof M. Mechanism-based pharmacokinetic-pharmacodynamic modeling of concentration-dependent hysteresis and biphasic electroencephalogram effects of alphaxalone in rats. J Pharmacol Exp Ther. 2002; 302(3): 1158-1167.
[6] Gibiansky L, Gibiansky E, Kakkar T, Ma P. Approximations of the target-mediated drug disposition model and identifiability of model parameters. J Pharmacokinet Pharmacodyn. 2008; 35(5): 573-591.
[7] Sharma A, Jusko WJ. Characteristics of indirect pharmacodynamic models and applications to clinical drug responses. Br J Clin Pharmacol. 1998; 45(3): 229–239.
[8] Karlsson MO, Beal SL, Sheiner LB. Three new residual error models for population PK/PD analyses. J Pharmacokinet Biopharm. 1995, 23(6): 651-672.
[9] Beal SL, Sheiner LB, Boeckmann A, Bauer RJ. NONMEM users guides. NONMEM Project Group, University of California, San Francisco, 1992.
Reference: PAGE 27 (2018) Abstr 8534 [www.page-meeting.org/?abstract=8534]
Poster: Drug/Disease Modelling - CNS