Integrated bioinformatics and interaction analysis to advance chronotherapies for mental disorders

Abstract

Introduction: Robust connections have been identified between the pathophysiology of mental disorders and the functioning of the circadian system. The overarching objective of this study was to investigate the potential for circadian rhythms to be leveraged for therapeutics in mental disorders.

Methods: We considered two approaches to chronotherapy-optimal timing of existing medications ("clocking the drugs") and redressing circadian abnormalities with small molecules ("drugging the clock"). We assessed whether circadian rhythm-modulating compounds can interact with the prominent drug targets of mental disorders utilizing computational tools like molecular docking and molecular dynamics simulation analysis.

Results: Firstly, an analysis of transcript-level rhythmic patterns in recognized drug targets for mental disorders found that 24-hour rhythmic patterns were measurable in 54.4% of targets in mice and 35.2% in humans. We also identified several drug receptors exhibiting 24-hour rhythmicity involved in critical physiological pathways for neural signaling and communication, such as neuroactive ligand-receptor interaction, calcium signaling pathway, cAMP signaling pathway, and dopaminergic and cholinergic synapses. These findings advocate that further research into the timing of drug administration in mental disorders is urgently required. We observed that many pharmacological modulators of mammalian circadian rhythms, including KL001, SR8278, SR9009, Nobiletin, and MLN4924, exhibit stable binding with psychotropic drug targets.

Discussion: These findings suggest that circadian clock-modulating pharmacologically active small molecules could be investigated further for repurposing in the treatment of mood disorders. In summary, the present analyses indicate the potential of chronotherapeutic approaches to mental disorder pharmacotherapy and specify the need for future circadian rhythm-oriented clinical research.

Keywords: chronotherapeutics; circadian rhythm; interaction analysis; mental disorders; molecular dynamics simulation.

FIGURE 1

Schematic illustration of the workflow designed to investigate (A) the dosing time-dependency of drugs for mental disorders and (B) interactions of pharmacological modulators of the circadian clock with the established therapeutic targets for mental disorders. The major steps in workflow A were identifying the targets of FDA-approved medications for prevalent mental disorders, analyzing the rhythmic patterns of these targets alongside their plasma half-lives, and assessing their functional relevance through gene ontology analysis. This investigation aimed to elucidate how the rhythmicity patterns of drug targets might influence drug effectiveness, particularly in drugs with shorter plasma half-lives. Workflow B explored whether pharmacological modulators of the circadian system could interact with the known targets of psychotropic drugs using molecular docking and MD simulation analysis. This investigation aimed to assess the potential repurposing of clock-modulating pharmacologically active compounds for the treatment of mental disorders. Created with BioRender.com.

FIGURE 2

Molecular targets for mental disorder drugs display robust daily rhythms. (A) Venn diagram showing overlapping rhythmic targets of drugs routinely used for mental and neurological disorders in mice and humans (Period 24 ± 3 h, JTK Q < 0.1). Information regarding the transcript-level rhythmic expression patterns of the targets is obtained from the CircaDB database (http://circadb.hogeneschlab.org/). (B) Multiple drugs share common 24-h rhythmic drug targets. (C) Distribution of rhythmic drug targets in different mental and neurological disorders. (D) Pie chart illustrating the distribution of plasma half-lives for mental disorder drugs with rhythmic targets. Drugs with plasma half-lives of less than 15 h and robustly rhythmic targets are most likely influenced by the dosing time (See Supplementary Table S1 for details). (E) Transcript-level rhythmic (Period±24 h, JTK Q < 0.1) expression patterns of representative drug targets for the four most prevalent mental disorders. The X-axis indicates time (h), and the Y-axis designates normalized intensity.

FIGURE 3

Rhythmic targets for mental disorder drugs are associated with diverse essential neurobiological processes and physiological pathways. (A) Categorizing rhythmic drug targets into functional groups by integrating information from the DAVID database (https://david.ncifcrf.gov/). Overrepresented biological processes, molecular functions, and cellular components are defined for the rhythmic drug targets (FDR <0.1). The top five statistically enriched (FDR <0.01, enrichment factor >1.5, count >5) GO terms from each category are depicted (See Supplementary Table S3–5 for details). (B) Physiological pathways associated with the molecular targets of mental disorder drugs. Pathways were obtained from the DAVID database (KEGG pathways). The top 20 statistically enriched (FDR <0.05, enrichment factor >1.5, count >5) canonical pathways are shown (see Supplementary Table S6 for details). (C) We observed that many rhythmic drug targets are associated with multiple physiological pathways. A polar plot depicted the highly widespread rhythmic drug targets involved in more than ten physiological pathways.

FIGURE 4

Circadian regulation of drug response in mental disorders. The figure visually represents the circadian regulations of the components of five key signaling pathways of the brain that are frequently targeted by the majority of mental disorder drugs (FDR <0.05, fold enrichment >2.5, count >5). The rhythmicity symbols specify the transcript level rhythmicity (JTK Q < 0.1, period 24 ± 3 h) of that component in the particular signaling pathway. (A)The neuroactive ligand-receptor interaction pathway encompasses G protein-coupled receptors (GPCRs) for neurotransmitters such as dopamine, serotonin, GABA, and acetylcholine, along with downstream effectors like calcium ion channels, cAMP, cGMP, PKA, and PKG. The downstream signaling cascades are displayed in panels (B–E), which play crucial roles in the pathophysiology of various psychiatric disorders by affecting gene expression, neuronal excitability, responses, and synaptic plasticity (Perez and Tardito, 2001; Reierson et al., 2011; Nakao et al., 2021). These pathways are closely intertwined and form a complex signaling network. A partial view of these interconnected pathways encompassing several rhythmic drug targets is depicted here. The complete pathway diagrams with rhythmic drug targets are shown in Supplementary Figure S5. Signaling pathways impacted by mental disorder drugs were retrieved from the DAVID database (https://david.ncifcrf.gov/). The rhythmicity of each component involved was investigated using the CircaDB database (http://circadb.hogeneschlab.org/). Created with BioRender.com.

FIGURE 5

Stable interactions between the pharmacological modulators of the circadian system and the established therapeutic targets of mental disorders (A) Binding free energy comparison of the 3D structure of pharmacological circadian modulators used for MD simulation analysis with molecular targets of lithium (GSK3β; PDB ID: 5K5N), valproic acid (PPARα; PDB ID: 1I7G), and melatonin (MTNR1A; PDB ID: 6ME3). (B) Binding models and interaction sites of PPARα, GSK3β, and MTNR1A with clock modulators exhibiting best binding affinity and stabilization within the receptor binding pocket (See Supplementary Table S10 for details). (C) Snapshots are shown at the beginning (left) and end (right) of the simulation for the best interaction models among PPARα, GSK3β, and MTNR1A with their ligands (clock modulators). (D) Root Mean Square Deviation (RMSD) and Root Mean Square Fluctuation (RMSF) values for the most robust interactions (PPARα-KL001, GSK3β-Nobiletin, and MTNR1A-SR8278) are displayed as a function of simulation time.