**1. Introduction**

Depression is the most common mental illness, affecting roughly 322 million people worldwide [1]. Depression is the main cause of disability and the fourth major contributor to the global illness burden [2]. Antidepressants are the third most commonly sold class of therapeutic drugs worldwide [3]. The majority of these treatments are based on chemicals that target the serotonin (5-hydroxytryptamine (5-HT): a group of G protein-coupled receptor and ligand-gated ion channels found in the central and peripheral nervous systems) transporter, a single protein in the brain. Selected serotonin reuptake inhibitors (SSRIs), which block 5-HT reuptake, account for around 80% of all antidepressants on the market [3]. Other antidepressants, such as serotonin and noradrenaline reuptake inhibitors, as well as traditional tricyclic antidepressants (e.g., amitryptyline, clomipramine, imipramine), prevent noradrenaline reuptake. Indeed, compared to tricyclic medicines, the success of selective serotonin reuptake inhibitors is mostly due to their safety, tolerability, and lack of severe side effects, which enhances patient compliance and quality of life [3].

Although seproxetine (SRX, also known as S-norfluoxetine) is classified as a selective serotonin reuptake inhibitor, its inhibitory action extends beyond serotonin transporters to dopamine transporters (DAT) and 5-HT2A/2C receptors [4]. It is the active N-demethylate metabolite of the commonly prescribed antidepressant fluoxetine and is deemed more potent than the parental compound itself [5]. The 5-HT(2A) and 5-HT(2C) receptors belong to the G-protein-coupled receptor (GPCR) superfamily. GPCRs interact with G-proteins to transmit extracellular signals to the inside of cells. The 5-HT(2A) and 5-HT(2C) receptors are involved in the effects of a wide range of drugs on anxiety, sleep patterns, depression, hallucinations, schizophrenia, dysthymia, eating behavior, and neuro-endocrine processes [6].

As SRX was found to be a 20 times more potent serotonin inhibitor than its sister enantiomer R-norfluoxetine, significant research efforts were focused on this drug in the 1990s [7]. However, serious cardiac side effects, such as QT prolongation (a measure of delayed ventricular repolarisation), halted further development [4,8]. The potency of SRX as a serotonin inhibitor should not be ignored, and an effort must be taken to chemically modify (charge–transfer complexation) SRX for a better serotonin inhibitor while suppressing the drawback.

Charge–transfer (CT) complexation, or electron–donor transfer, is a crucial aspect of biochemical and biological processes such as drug design, enzyme catalysis, and ion sensing [9]. The pharmacodynamics and thermodynamics of therapeutic substances and biological processes in the human body are studied using charge–transfer complexation interactions [10–14]. In biological systems, charge–transfer complexes may play a crucial function. Extensive research has been carried out on charge–transfer interactions between inorganic anions, particularly the iodide ion and pyridinium, and substituted pyridinium cations, to determine the sensitivity of their charge–transfer absorption to the solvent environment, as well as the potential role of structures of this type in enzymatic oxidationreduction processes [15]. As the charge–transfer complexes are a simpler, cheaper, and more efficient tool of analysis than the other methods mentioned in the literature, charge– transfer interactions are an important subject employed in the determination of medicines in pharmaceutical and pure forms [16].

Many reports stated the interactions, in solution, between flavin mononucleotide, flavin adenine dinucleotide, or riboflavin and a variety of donors, including hydrocarbons [17], indoles [18], NADH [19], NADPH [19], purines and pyrimidines, as well as other compounds with no obvious donor properties. There is little doubt that complete electron transfer happens in several of these systems to generate the flavin semiquinone [20]. The new broad absorption band reported for mixes of the reduced form of flavin mononucleotide (FMNH2) and (FMN) was attributed to the creation of charge–transfer complexes [21]. 2-methyl-1,4-naphthoquinone, also known as vitamin K3, used as a synthetic substitute for K1, o-quinone adrenochrome, and many other biologically important quinones have substantial electron donor complexing capacity [22].

Tryptophan appears to be unique among amino acids in its capacity to generate charge transfer complexes due to the strong donor characteristics of the indole ring. However, another study has shown that a pyridinium model compound of NAD+ may form complexes with tyrosine and phenylalanine [23]. Spectral evidence was also found to produce charge–transfer complexes between NAD+ and model pyridinium compounds with chymotrypsinogen, a tryptophan-rich protein [24]. chymotrypsinogen, a tryptophan-rich protein [24]. Molecular docking (MD) is a computer method for efficiently predicting the non-covalent binding of macromolecules (receptors) and small molecules (acceptors) based on their unbound structures, structures generated through MD simulations, homology modeling, and other methods. The prediction of small molecule binding to proteins is of particular practical significance since it is used to screen virtual libraries of drug-like com-

Tryptophan appears to be unique among amino acids in its capacity to generate charge transfer complexes due to the strong donor characteristics of the indole ring. However, another study has shown that a pyridinium model compound of NAD+ may form complexes with tyrosine and phenylalanine [23]. Spectral evidence was also found to produce charge–transfer complexes between NAD+ and model pyridinium compounds with

*Molecules* **2022**, *27*, x 3 of 21

Molecular docking (MD) is a computer method for efficiently predicting the noncovalent binding of macromolecules (receptors) and small molecules (acceptors) based on their unbound structures, structures generated through MD simulations, homology modeling, and other methods. The prediction of small molecule binding to proteins is of particular practical significance since it is used to screen virtual libraries of drug-like compounds for leads for further drug development. As a result, MD has become an important method in drug development. pounds for leads for further drug development. As a result, MD has become an important method in drug development. Here, we used the Autodock Vina program to investigate the interactions between the ligand (SRX and synthesized CT complexes) and receptors (serotonin, dopamine, and TrkB kinase receptors). In the 1970s and 1980s periods, selective serotonin reuptake inhibitors (SSRIs) were developed, which are as effective antidepressants as tricyclics but do not have as many side effects as other antidepressant drugs. Binding energy, along with

Here, we used the Autodock Vina program to investigate the interactions between the ligand (SRX and synthesized CT complexes) and receptors (serotonin, dopamine, and TrkB kinase receptors). In the 1970s and 1980s periods, selective serotonin reuptake inhibitors (SSRIs) were developed, which are as effective antidepressants as tricyclics but do not have as many side effects as other antidepressant drugs. Binding energy, along with hydrophobic properties, ionizability, aromatic, and hydrogen bond surfaces, were also investigated. The molecular dynamic simulation was achieved at 300 K for 100 ns. The dynamic properties of the complexes were compared in many characterizations such as residue flexibility, structural solidity, solvent-accessible surface area, and other measurements. DFT using the B-3LYP/6-311G++ (basis set) level of theory was employed to obtain an optimized geometry of the CT complex- [(SRX)(PA}], [(SRX)(DNB)], [(SRX)(p-NBA)], [(SRX)(DCQ)], [(SRX)(DBQ)], and [(SRX)(TCNQ)] with minimal energy. Different parameters of the complexes were obtained and compared. hydrophobic properties, ionizability, aromatic, and hydrogen bond surfaces, were also investigated. The molecular dynamic simulation was achieved at 300 K for 100 ns. The dynamic properties of the complexes were compared in many characterizations such as residue flexibility, structural solidity, solvent-accessible surface area, and other measurements. DFT using the B-3LYP/6-311G++ (basis set) level of theory was employed to obtain an optimized geometry of the CT complex- [(SRX)(PA}], [(SRX)(DNB)], [(SRX)(p-NBA)], [(SRX)(DCQ)], [(SRX)(DBQ)], and [(SRX)(TCNQ)] with minimal energy. Different parameters of the complexes were obtained and compared. **2. Materials and Methods**  *2.1. Synthesis of [(SRX)(π-Acceptor)] Charge–Transfer Complexes*  The charge–transfer complexes [(SRX)(π-acceptor)] where π-acceptor are PA, DNB,

### **2. Materials and Methods** *p*-NBA, DCQ, DBQ, and TCNQ (Figure 1) were synthesized as 1:1 by the reaction of SRX

### *2.1. Synthesis of [(SRX)(π-Acceptor)] Charge–Transfer Complexes* donor in a solution (25 mL) of each acceptor [25].

The charge–transfer complexes [(SRX)(π-acceptor)] where π-acceptor are PA, DNB, *p*-NBA, DCQ, DBQ, and TCNQ (Figure 1) were synthesized as 1:1 by the reaction of SRX donor in a solution (25 mL) of each acceptor [25]. At room temperature, the mixtures were agitated for about an hour in each case. The precipitate was filtered and washed with the smallest amount of dichloromethane possible before being dried under vacuum over anhydrous CaCl2.

**Figure 1.** Speculated molecular structures of (1:1) charge-transfer complexes [(SRX)(π-acceptor)]. **Figure 1.** Speculated molecular structures of (1:1) charge-transfer complexes [(SRX)(π-acceptor)].

At room temperature, the mixtures were agitated for about an hour in each case. The precipitate was filtered and washed with the smallest amount of dichloromethane possible before being dried under vacuum over anhydrous CaCl2.

### *2.2. Instruments and Measurements 2.2. Instruments and Measurements*

With safeguards (platinum pans, nitrogen gas flow, and 30 ◦C min−<sup>1</sup> heating rate), thermogravimetric analysis (TGA/DTG) was examined using Shimadzu TGA-50H equipment. A Perkin–Elmer Precisely Lambda 25 UV/Vis Spectrometer was used to scan the electronic absorption spectra of the synthesized charge–transfer complexes in the 200–800 nm region. A Bruker 600 MHz spectrometer was used to measure <sup>1</sup>H-NMR spectra in DMSO solvent. With safeguards (platinum pans, nitrogen gas flow, and 30 °C min−1 heating rate), thermogravimetric analysis (TGA/DTG) was examined using Shimadzu TGA-50H equipment. A Perkin–Elmer Precisely Lambda 25 UV/Vis Spectrometer was used to scan the electronic absorption spectra of the synthesized charge–transfer complexes in the 200–800 nm region. A Bruker 600 MHz spectrometer was used to measure 1H-NMR spectra in

### *2.3. Molecular Docking* DMSO solvent.

The structures of the SRX drug and CT complexes were handled in PDBQT format via OpenBabelIGUI software (version 2.4.1) [26]. Then, the PyRx-Python prescription 0.8 and MMFF94 force field were used to minimize the energy of the structure for 500 steps [27]. The RCSB Protein Data Bank [28] was used to get the 3D crystal structures of the three receptors. The receptors were arranged using the BIOVIA Discovery Studio Visualizer (v19.1.0.18287). Kollman charges were also measured with the help of the AutoDock Tool [29]. The Geistenger method was used to allocate partial charges. The docking calculations were performed with Autodock Vina [30]. The DS (Discovery Studio) Visualizer was used to examine the docked poses that resulted. *2.3. Molecular Docking*  The structures of the SRX drug and CT complexes were handled in PDBQT format via OpenBabelIGUI software (version 2.4.1) [26]. Then, the PyRx-Python prescription 0.8 and MMFF94 force field were used to minimize the energy of the structure for 500 steps [27]. The RCSB Protein Data Bank [28] was used to get the 3D crystal structures of the three receptors. The receptors were arranged using the BIOVIA Discovery Studio Visualizer (v19.1.0.18287). Kollman charges were also measured with the help of the AutoDock Tool [29]. The Geistenger method was used to allocate partial charges. The docking calculations were performed with Autodock Vina [30]. The DS (Discovery Studio) Visualizer
