Modulation of serotonin transporter expression by escitalopram under inflammation

Abstract

Selective serotonin reuptake inhibitors (SSRIs) are widely used for depression based on the monoamine deficiency hypothesis. However, the clinical use of these agents is controversial, in part because of their variable clinical efficacy and in part because of their delayed onset of action. Because of the complexities involved in replicating human disease and clinical dosing in animal models, the scientific community has not reached a consensus on the reasons for these phenomena. In this work, we create a theoretical hippocampal model incorporating escitalopram's pharmacokinetics, pharmacodynamics (competitive and non-competitive inhibition, and serotonin transporter (SERT) internalization), inflammation, and receptor dynamics. With this model, we simulate chronic oral escitalopram in mice showing that days to weeks are needed for serotonin levels to reach steady-state. We show escitalopram's chemical efficacy is diminished under inflammation. Our model thus offers mechanisms for how chronic escitalopram affects brain serotonin, emphasizing the importance of optimized dose and time for future antidepressant discoveries.

Fig. 1. Modeling evoked FSCV antidepressant data.

a Schematic of experimental protocol for evoked FSCV serotonin signals in the CA2 region of the hippocampus. b Representative FSCV color plot control file. Inset: The vertical trace (i vs. v) shows a representative serotonin CV, while the horizontal trace (i vs. t) shows the release and reuptake profile of serotonin over the course of the experiment. c Experimental concentration vs. time traces from the CA2 region of the hippocampus of female mice (n = 4 animals each dose, mean ± SEM) pre- and post-escitalopram intraperitoneal administration at time and dose given in the legend. d Simulation of serotonin extracellular concentration pre- and post-escitalopram at time and dose given in the legend solely assuming competitive inhibition. e Simulation following the same protocol as in D, adding non-competitive inhibition. f Simulation following the same protocol as in E and adding SERT internalization. Illustrations made with Biorender.com.

Fig. 2. Representation of the relationships in the mathematical model of serotonin, histamine, and glia.

a Schematic of serotonin and histamine neuronal terminals. b Schematic of glia. c Detailed modeled SERT processes. Symbols of the model’s main variables are depicted in red rectangular labels and described in Table 1. The acronyms of main enzymes, transporters, and receptors are represented with blue elliptic labels and described in Table 2. The pharmacological effects of escitalopram (ESCIT) and α-fluoromethylhistidine (FMH) are represented in purple rectangular labels. Notice that variables and enzymes with the same name in the schematic but present in different compartments are different entities (e.g., HTDC in histamine neurons and HTDC in glia). An extended description of each variable and enzyme/transporter by compartments can be found in the Supplementary Information. Illustration made with LaTeX Tikz pachage.

Fig. 3. Model of oral chronic dosing with escitalopram.

a Oral dosing paradigm, (i) shows the differences in escitalopram time profiles in the brain after i.p. and oral administration, (ii) shows the increase of escitalopram when the administration is repeated before clearance, and (iii) shows how escitalopram reaches a steady-state after several administrations. b Modeling of brain concentration of escitalopram (i), serotonin (ii), and SERT surface ratio (iii) following an i.p. injection or oral administration of 1.04 mg kg⁻¹ (equivalent to 5 mg human dose, see Methods). Half-lives of the clearance are provided in the panels. c Modeling oral chronic dosing effects on escitalopram (i), serotonin (ii), and SERT surface ratio (iii). Administration is repeated every 8 h (equivalent to human daily dose, see Methods section). Maximum and minimum values of oscillations, as well as the mean accumulated concentration is given in the panels. Results for the 10 mg and 20 mg human doses can be found in the Supplementary Information Information (Figs. S2, S3). Illustrations made with Biorender.com.

Fig. 4. Inflammation effects on oral chronic escitalopram efficacy.

a Mast cell model of histamine degranulation. In (i), the schematic of the computational model is given. The main variables are depicted in red rectangular labels. The acronyms of main enzymes, transporters, and receptors are represented with blue elliptic labels. Pharmacological effects are represented in purple rectangular labels. In (ii), the modeled reaction of histamine (purple) and serotonin (red) to an inflammation trigger. b Modeling of oral chronic dosing escitalopram (5 mg pill, ~1.02 mg kg⁻¹) effects on brain escitalopram (i), serotonin (ii), and SERT surface ratio (iii) during control state and inflammation. For the inflammation simulation, mast cell and glia production and release of histamine is triggered 35 days before the first dose. Administration is then repeated every 8 h. c Modeling of extracellular histamine (i) and serotonin (ii) following 20 mg kg⁻¹ i.p. injection of (S) α-fluoromethylhistidine 1 h after the start of the simulation. d Modeling of oral chronic co-administration of FMH (2.5 mg analogous dose for mice, ~0.51 mg kg⁻¹) and escitalopram dosing effects on brain FMH (i), serotonin (ii), and SERT surface ratio (iii) during inflammation (purple) and comparison to control administration of escitalopram as described on panel B (blue). For the inflammation simulation, mast cell and glia production and release of histamine is triggered 35 days before the first dose. Administration is then repeated every 8 h. Results for the 10 mg and 20 mg human doses can be found in the Supplementary Information (Figs. S4,S5). Illustrations made with Biorender.com.