A Systems Pharmacology model of peripheral serotonin production proposing two different synthesis compartments with markedly different release rates into blood
M. Machacek (1), L. Renaud (1), C. Kohl (2), M. Vercauteren (2), L. Remen (2), R. Welford (2)
(1) LYO-X GmbH, Ringstrasse 9, 4123 Allschwil, Switzerland. (2) Idorsia Pharmaceuticals Limited, Hegenheimermattweg 91, 4123 Allschwil, Switzerland
Objectives: High levels of systemic serotonin (5-HT) are implicated in several diseases including carcinoid syndrome, pulmonary arterial hypertension and obesity. Systemic 5-HT is produced from dietary tryptophan (Trp) where the rate limiting step its catalysis by tryptophan hydroxylase 1 (TPH1). A TPH1 enzyme inhibitor was recently approved for treatment of carcinoid syndrome and others are in development. To enable observation of changes in 5-HT synthesis in vivo on an hour time scale and to test new TPH1 inhibitors we previously described an approach utilizing a stable isotope Trp tracer (h-Trp) [1]. The aim of the current study was to develop a Systems Pharmacology model to explain the kinetics, distribution and compartmentalization involved in h-5-HT and 5-HT production and to utilize the model to predict from single dose experiments long term effects on 5-HT with chronic dosing of TPH1 inhibitors.
Methods: Data from a h-Trp study in Beagle dog was used to develop a pharmacokinetic (PK)/ pharmacodynamic (PD) model. Dogs received a single oral dose of 12 mg/kg h-Trp one hour after receiving orally vehicle, 0.2, 1 or 5 mg/kg of a TPH1 inhibitor. TPH1 inhibitor PK, h-Trp, Trp, h-5-HT and 5-HT observations were available for 1 day, and 5-HT pathway observations in the vehicle group for 11 days. As expected for a single dose treatment, no effect on whole blood 5-HT was observed due to its slow turnover. A compartmental population PK/PD approach was used to describe the concentration-time curves of the enzyme inhibitor in plasma, and h-Trp, Trp, h-5-HT and 5-HT in whole blood. It was assumed that the distribution parameters were identical for the isotope labeled and non-labeled species. Further, the enzymatic reaction rates were assumed to be identical and the competition between Trp and h-Trp for the enzyme TPH1 was modelled as described in [2]. The simples t compartment structure was sought that would explain the observed data and the model parameters were estimated with Monolix 4.3.3.
Results: Trp had a three-compartmental kinetics with the first two distribution half-lives of 6.5 minutes and 7.6 hours, and a terminal half-life of 5.8 days. The conversion of Trp to 5-HT occurred in two different peripheral effect compartments with two very different exchange rate constants for 5-HT with the plasma compartment. These two reaction compartments were different from the two peripheral compartments for the distribution of Trp and simpler models were unable to describe the observed h-Trp and h-5-HT. In the first compartment, h-5-HT had a half-life equivalent of 6.7 seconds and in the second compartment of 16.9 hours. Thus, after oral intake of h-Trp there was immediate appearance of h-5-HT in the circulation because of the fast compartment; while the slow compartment was responsible for the storage and release of h-5-HT into the circulation after h-Trp has been eliminated. The half-life of whole blood 5-HT in dog was estimated as 4.2 days. Plate lets, that are the main site of 5-HT storage in blood have a half-life of 3.3 days [3]. From the model, it was found that 90% of the 5-HT is in blood, 10 % in the slow storage compartment and 0.001% in the fast storage compartment. For model validation, the effect of daily dosing of 30 mg/kg of the TPH1 test inhibitor administration for 2 weeks on whole blood 5-HT in dogs was successfully predicted. Further, the model correctly predicted the blood Trp levels in dog after 24 hours fasting followed by daily feeding for 11 days.
Conclusions: The model built from the stable isotope tracing data indicated the need for slow and fast 5-HT production compartments. The fast release compartment is likely the gut, the major site of peripheral 5-HT synthesis. The second compartment may represent synthesis in other organs such as lung [4] or a slower release process from gut at a different time scale. 90% of the 5-HT is stored in platelets and only 10% at the site where it is synthesized. Different sites of peripheral 5-HT synthesis and different exchange rates will be critically relevant to the design and development of novel effective TPH1 inhibitors. Further, the System Pharmacology model can be further developed as a translational tool to predict doses of TPH1 inhibitors to reach different 5-HT lowering thresholds in patients.
References:
[1] Richard W. D. Welford, Magali Vercauteren, Annette Trébaul, Christophe Cattaneo, Doriane Eckert, Marco Garzotti, Patrick Sieber, Jérôme Segrestaa, Rolf Studer, Peter M. A. Groenen & Oliver Nayler. Serotonin biosynthesis as a predictive marker of serotonin pharmacodynamics and disease induced Dysregulation. Scientific Reports | 6:30059 | DOI: 10.1038/srep30059
[2] T. Pocklington, J. Jeffery. Competition of two substrates for a single enzyme. A simple kinetic theorem exemplified by a hydroxy steroid dehydrogenase reaction. Biochemical Journal (1969) 112 (3) 331-334.
[3] Ryou Tanaka, Ayako Murota, Yukiko Nagashima, and Yoshihisa Yamane. Changes in Platelet Life Span in Dogs with Mitral Valve Regurgitation. J Vet Intern Med 2002;16:446–451.
[4] Rabih El-Merahbi, Mona Löffler, Alexander Mayer, Grzegorz Sumara. The roles of peripheral serotonin in metabolic homeostasis. FEBS Letters 589 (2015 ) 1728–1734.