Next Article in Journal
The Role of Tetrahydrocurcumin in Tumor and Neurodegenerative Diseases Through Anti-Inflammatory Effects
Previous Article in Journal
Environmental Factors Exacerbate Parkinsonian Phenotypes in an Asian-Specific Knock-In LRRK2 Risk Variant in Mice
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
IJMS
Volume 26
Issue 8
10.3390/ijms26083559
ijms-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

2-Phenylcyclopropylmethylamine (PCPMA) Derivatives as D3R-Selective Ligands for 3D-QSAR, Docking and Molecular Dynamics Simulation Studies

by
Li Guo
1,
Yuepeng Gao
1,
Sujuan Zhang
1,
Lingmi Zhao
1,
Runxin Zhao
1,
Pinghua Sun
1,2,
Xinhui Pan
1,3,* and
Wei Zhang
1,*
1
Key Laboratory of Xinjiang Phytomedicine Resource and Utilization Ministry of Education, School of Pharmacy, Institute for Safflower Industry Research, Shihezi University, Shihezi 832002, China
2
International Cooperative Laboratory of Traditional Chinese Medicine Modernization, Innovative Drug Development of Chinese Ministry of Education (MOE), College of Pharmacy, Jinan University, Guangzhou 510632, China
3
Stake Key Laboratory of Natural and Biomimetic Drugs, Department of Chemical Biology, School of Pharmaceutical Sciences, Peking University, Beijing 100191, China
*
Authors to whom correspondence should be addressed.
Int. J. Mol. Sci. 2025, 26(8), 3559; https://doi.org/10.3390/ijms26083559
Submission received: 12 March 2025 / Revised: 4 April 2025 / Accepted: 6 April 2025 / Published: 10 April 2025
(This article belongs to the Section Molecular Informatics)
Browse Figures
Versions Notes

Abstract

Dopamine D3 receptor (D3R) is a key receptor for regulating motor, cognitive, and other functions. In this study, 50 2-phenylcyclopropylmethylamine (PCPMA) derivatives with good selectivity for D3R were investigated using a three-dimensional quantitative structure–activity relationship (3D-QSAR) method. The CoMFA and CoMSIA model results showed good predictive ability, as evidenced by high r2 and q2 values. 3D-QSAR results showed that steric, electrostatic, and hydrophobic fields played important roles in the binding of PCPMAs to D3R. Based on above results, four novel PCPMAs were designed, which were predicted to have a stronger affinity with D3R. Molecular docking combined with 300 ns molecular dynamics simulations were performed to reveal the mode of interaction between D3R and PCPMAs. Additionally, a combination of free energy calculations and energy decomposition results indicated strong interaction between the ligands and residues in the binding pocket of the D3 receptor. This work provides suggestions for exploring more selective D3R ligands, and this theoretical framework also lays the foundation for future experimental investigations to evaluate the pharmacological characteristics and binding affinities of novel derivatives.

1. Introduction

Dopamine (DA) receptors are widely distributed in the brain and are one of the most important neurotransmitters in the nervous system. These receptors are G protein-coupled receptors (GPCRs) that play vital roles in various functions such as motor control, motivation, and cognitive processes. They have been classified into two subfamilies, D1-like (D1 and D5 receptors) and D2-like (D2, D3, and D4 receptor subtypes), on the basis of sequence differences, activation of G-protein-coupled receptors, and pharmacological properties. The D1-like receptors activate a stimulatory G protein-coupled receptor to increase adenylate cyclase activity, while D2-like receptors activate inhibitory G protein alpha subunit (Gi/o)-coupled receptors to inhibit adenylate cyclase and modulate other effector pathways [1,2,3]. D2 and D3 receptors (D2R and D3R), belonging to the D2-like category, are particularly important as therapeutic targets for neuropsychological disorders like schizophrenia, Parkinson’s disease, and drug addiction. Because of the high degree of sequence identity (78%) within the transmembrane (TM) domains and the close identity of the residues that form the binding site with the receptor, this makes it challenging to create subtype-selective agents with physicochemical properties suitable for in vivo characterization of the physiological actions of these receptor subtypes [4]. D2R is the primary target for traditional antipsychotic medications, but the presence of antagonism with the cerebral striatum has led to the emergence of side effects, including extrapyramidal symptoms (EPS) [3], elevated prolactin, drug abuse, and damaging effects on motivated behavior. D3R has been a therapeutic target of great interest due to its relatively localized expression in the mid-limbic neural circuits, including the nucleus accumbens, hippocampal island of Calleja, ventral striatum, and the selective distribution of D3R in the nucleus ambiguus. In particular, it has the potential to reduce the incidence of side effects such as EPS and hyperprolactinemia, which has enhanced the interest in D3R as a potential target for drug addiction [5]. As a result, there is a pressing need for the design and development of highly selective D3R ligands to address psychiatric disorders [6,7,8].
D3R-selective ligands in clinic and clinical trials are shown in Figure 1A. SB277011A, introduced in 2000 [9], has become an essential scaffold in the study of dopamine D3 receptors (D3R). As a competitive D3R-selective antagonist, it had been employed extensively in vivo to elucidate the role of D3R in various neural circuits and its potential for therapeutic applications in neuropsychiatric disorders. In 2010, Micheli et al. identified GSK598809 as a D3R antagonist, notable for its exceptional 100-fold selectivity for D3R over D2R. This compound belonged to a new 1,2,4-triazol-3-yl-azabicyclo[3.1.0]hexanes backbone with a favorable pharmacokinetic profile, but a potential cardiovascular risk was associated with high doses of GSK598809 in clinical trials [10]. Buspirone, a D3R-preferring 5-HT 1A antagonist, but it was noteworthy that its selectivity for D3R was only observed with oral administration, suggesting a relationship with D3R that could be attributed to the metabolism of buspirone. It appeared that the active fraction contributing to its D3R selectivity may be a metabolite generated during its first pass through the liver [11]. RGH188 (cariprazine, CRP) was also a D3R-preferring agent, showing a 10-fold selectivity for D3R over D2R. Noteworthily, cariprazine was shown to have cognitive enhancement in vivo, further underscoring the therapeutic potential of D3R-selective ligands in treating cognitive deficits common in neuropsychiatric disorders [12].
Despite progress in developing highly selective D3R ligands, challenges such as low bioavailability, poor brain permeability, and adverse side effects observed in clinical trials remain [8,10]. These issues emphasize the necessity for continued research targeting D3R for drug development, as effective and safe therapeutic options are still needed for neuropsychiatric disorders. Currently, the design of D3R-selective ligands is mostly based on the strategy of using the unique characteristics of the D3R binding pocket to create new compounds that can engage in multi-faceted interactions, potentially leading to enhanced selectivity and affinity [10,13,14]. Inspired by fragmentation and molecular docking, Cheon et al. designed and synthesized a series of piperazine–phthalimide bitopic ligands, and found that compound 9i exhibited a high D3R selectivity (100-fold higher than D2R) and 199-fold increase in agonist activity compared to quinpyrazole [15]. Shaik et al. employed the crystal structure of the D3R complexed with eticlopride to design a novel bitopic ligand, and highlighted that the pyrrolidine ring O-alkylated derivatives showed high affinity with D2R/D3R [16]. Francisco O Battiti et al. found that FOB02-04A is a bitopic and full agonist of D3R. Meanwhile, a selectivity site was revealed by determining the cryo-electron microscopy (cryo-EM) structure of the hD3R:GαOβγ complex bound to the D3R selective bitopic agonist FOB02-04A [17]. Besides, PCPMA derivatives have been reported as serotonin transporter inhibitors [18], melatonin receptor agonists [19], 5-HT2C receptor agonists [20], etc, which have potential value in the treatment of neurological disorders such as schizophrenia and addiction [21,22]. Recently, a series of PCPMAs, particularly compounds (1R,2R)-22e and its enantiomer (1S,2S)-22e, were designed as selective D3R ligands. Among them, compound (1R,2R)-30q showed a significant affinity for D3R with a Ki value of 2.2 nM in mice. These compounds were strategically designed by incorporating a linker that enhanced binding to D3R and a moiety that bound to the extended binding pocket (EBP) of the receptor [23].
As a 5-HT 2C receptor agonist, a PCPMA derivative modified by the virtual screening lead compound cyclopropylamine showed antidepressant-like effects in commonly used behavioral assays, similar to the potential therapeutic effect of D3R [20]. The clinical extended-release melatonin preparation tasimelteon containing a PCPMA backbone also had similar effects to D3R in neuroprotection and drug abuse by activating two high-affinity G protein-coupled receptors, Melatonin Receptor 1 (MT1) and Melatonin Receptor 2 (MT2) [19]. This highly selective PCPMA 5-HT 2C agonist compound also showed moderate binding affinity for dopamine D3R. Therefore, Chen et al. linked the junction moiety that binds well to D3R and the group that binds to the extended binding pocket (EBP) to the nitrogen atom of PCPMA, and the designed compounds showed high binding affinity for D3R, good selectivity, bioavailability, and brain penetration. Therefore, we selected these compounds for research in order to obtain D3R ligands with better selectivity and higher affinity [23].
3D quantitative structure–activity relationship (3D-QSAR) studies have been found to be of great significance in ligand-based drug design and potent drug development, which correlates 3D-structural features of the chemicals with their associated properties [24,25,26]. Comparative molecular field analysis (CoMFA) [27] and comparative molecular similarity index analysis (CoMSIA) [28] have been employed. The former correlated steric and electrostatic properties of the molecular structure with its activity, while the latter correlated the hydrophobic field, hydrogen-bond donor/hydrogen-bond acceptor and steric/electrostatic field with its activity.
Given the potential of PCPMAs as D3R ligands, an in-depth computational study of a series of PCPMAs was performed to explore their quantitative structure activity relationship and further improve their selectivity and affinity for D3R. To establish a robust dataset, the affinities (pKi) and molecular structures of 50 compounds reported by Cheng’s group were selected [23] (see Table S1). Based on the results of the 3D-QSAR modeling and contour map analysis, a series of novel derivatives were designed, and their binding affinities with D3R were predicted. The outcomes of this work were expected to enhance the understanding of the structure–activity relationships of PCPMA derivatives and provide a reliable framework for predicting the biological activity of new candidates. Taken together, our results contribute to the design of novel D3R ligands and provide a theoretical basis for the discovery of molecules with potential antipsychotic activity. The workflow of the current study is represented in Scheme 1.

2. Results and Discussion

2.1. CoMFA and CoMSIA Statistical Results

A 3D-QSAR study was performed on 50 PCPMA derivatives in this work. All molecules were listed in Table S1. The compounds in the training set were aligned to compound 30r, which had the highest affinity. It could be seen from Figure 2 that the common skeleton of all molecules overlapped. One CoMFA model and 32 CoMSIA models were developed by the 3D-QSAR study. The statistical parameters of CoMFA and the best CoMSIA models are listed in Table 1. The best CoMSIA model, combining steric and hydrophobic fields, and the CoMFA model were selected to predict the binding affinity of compounds with D3R. The predicted affinity (pKi) values of the training and test sets are shown in Table S1.
To give the 3D-QSAR model a stable and reliable prediction ability, the q2 of the constructed model was generally larger than 0.5, while r2 was close to 1. The smaller the SEE value, the better the quality. For the CoMFA model, q2 was 0.607, N was 10, r2 was 0.981, SEE was 0.094, and F was 149.222, respectively. In summary, the CoMFA model in this study was reliable, and the contributions of the steric and electrostatic fields were 39.3% and 60.7%, respectively, indicating that the electrostatic field had an important effect on the binding affinity of the PCPMA ligands with D3R. This may be consistent with the two-step mechanism of GPCR ligand binding—electrostatic interactions determine binding specificity (orientation), while hydrophobic interactions provide binding energy (stabilization) [29,30].
For the CoMSIA model, 32 models with different field combinations were obtained (see Table S2). The combination of electrostatic and steric fields was selected as the best CoMSIA model due to its q2 of 0.643, r2 of 0.889, and the presence of four components. These statistics showed that the developed CoMSIA model has good reliability. In this model, the contributions of the steric and hydrophobic fields on compound affinity were 37.5% and 72.5%, respectively, indicating that hydrophobic interactions had an important role in the binding affinity of PCPMA ligands with D3R. It is generally known that CoMSIA models tend to outperform CoMFA models, but in our study, CoMFA performed better in this case. The nature of the molecules in our dataset might be more suitable for the CoMFA approach. CoMFA focuses on steric and electrostatic fields, which could capture the key interactions relevant to our set of compounds more effectively than the CoMSIA model, perhaps because the steric and electrostatic effects dominate the binding mechanisms of our compounds. And our dataset is relatively small, so the CoMFA model might be more robust and less prone to overfitting compared to the more complex CoMSIA model [31,32,33].
The predicted values of pKi for all compounds in the test set, calculated using CoMFA and the best CoMSIA model, are shown in Table S1, and the difference between the experimental and predicted affinity was found to be small. The correlation between the predicted affinity of the models and the experimental affinity of the training and test sets was then verified using scatter plots of the regression curves. As shown in Figure 3, most of the points were evenly distributed on both sides of the regression curve, which further proves the reliability of the 3D-QSAR model.

2.2. CoMFA and CoMSIA Contour Map Analysis

To investigate the structure–activity relationship between various factors and compound structures in 3D-QSAR modeling, contour maps of CoMFA and optimal CoMSIA models were constructed based on template compound 30r (See Figure 4 and Figure 5). The contributions of favorable and unfavorable zones in all contour maps were 80% and 20%, respectively.
The CoMFA modeling results show that the binding affinity of PCPMA ligands with D3R was related to the spatial and electrostatic interactions of the ligands. The contour plots of the steric field are shown in Figure 4A, where green and yellow contours indicate that bulky groups were favorable and unfavorable for affinity, respectively. There was a large green contour near the phenyl and in the methylamine side chain, suggesting that the presence of bulky groups favored the activity. The contour plot of the electrostatic field is shown in Figure 4B, where the blue and red contours display whether positive charge was favorable or unfavorable for activity, respectively. There is a red contour at the end, attached to the amide bond of the methylamine linker chain, indicating that a negatively charged substitution could increase the affinity.
The best CoMSIA model results suggest that the magnitude of the affinity between the PCPMA ligands and the D3R was influenced by the molecular steric structure field and hydrophobic interactions. For the CoMSIA model, this mainly consists of a green region at the methylamine N side chain substitution, where the introduction of a bulky group may improve compound affinity, consistent with the CoMFA model; a large yellow region at the amide bond at the end of the fatty linkage chain and at the aryl substitution of the linked aryl group, suggesting that the affinity of the compound may be improved by the introduction of a smaller bulky group here; and finally a smaller green area at the aryl rear substituent, where the addition of a bulky group could be attempted to increase the affinity (shown in Figure 4A and Figure 5A). Furthermore, contour plots of the hydrophobic field are shown in Figure 5B, where yellow and grey contours indicate the favorable and unfavorable affinity of the hydrophobic groups, respectively. There is a large yellow contour at the vicinity of the phenyl group, indicating that the presence of the hydrophobic group was conducive to increasing affinity, and a large grey contour near the amide bond of the methylamine linkage chain, demonstrating that the presence of hydrophilic groups favored affinity.

2.3. Design of Novel PCPMA-Like Compounds

Based on the information obtained from the above contour plots, the key points of the subsequent structural modification of the compound 30r with the best affinity are summarized (see Figure 6). Firstly, the substituents on the benzene ring of the cyclopropyl side chain of the PCPMA ligands preferred bulky hydrophobic groups such as halogenated hydrocarbons, aryl rings, and so on. Secondly, the affinity of the PCPMA ligands with D3R was enhanced when bulky groups such as benzene, biphenyls, and amino acids were present on the N-propyl side chain. Lastly, the affinity of the compounds could be increased when the group attached to the amide bond was a negatively charged hydrophilic group.
Based on the above results, a series of novel PCPMA ligands were designed, as shown in Table 2. The designed compounds D1 to D4 exhibited good predicted affinities compared to template compound 30r, and more in-depth studies were conducted on these four compounds. The nitrogen-containing aliphatic linkage chain terminal amide bond connecting 1-phenyl-1-H-1,2,4-triazole and imidazole favored the compounds’ affinity (see compounds D1, D4 in Table 2), and the affinity of the compounds designed by the addition of nitrogen atoms between the nitrogen-containing aliphatic linkage chain terminal amide bond and 1-phenyl-1-H-1,2,4-triazole also increased (see compound D2 in Table 2). The activity of compounds with bulky groups such as biphenyl compounds was also increased by the addition of bulky groups to the side chains on the nitrogen atoms of the connecting chains (see compound D3 in Table 2). Comparable binding affinities between the newly designed compounds D1 to D4 require further experimental validation.

2.4. ADMET and Drug Similarity Prediction Results

The ADMET performance of the newly designed compounds and the template compound were predicted using the ADMET lab 3.0 online website [34], and the drug properties and synthetic accessibility scores of the new compounds were also evaluated. As shown in Table 3, the LogP of all the designed compounds were smaller, and the topological polar surface area (TPSA) was used to assess the solubility of the compounds, the blood–brain barrier permeability, and the protein binding; all compounds scored within the normal range. The synthesis accessibility scores of the newly designed compounds were all less than 6, suggesting that all designed compounds could easily to be synthesized. As shown in Table S2, the oral bioavailability (F30%), as the crucial parameter of absorption for orally administered drugs, was 0–0.1, indicating excellent oral bioavailability. As shown in Table S3, the plasma protein binding (PPB) and volume of distribution (Vd) were calculated to describe the distribution behavior; all designed compounds exhibited PPB values > 90%, indicating high levels of plasma protein binding and probable low therapeutic indexes. Vd values are a parameter characterizing the duration of drug retention in plasma or redistribution into other tissues [35], which were in the range 0.04–20.00 L/kg in our study; all compounds exhibited Vd values in the range of 0.40–1.00 L/kg. Concerning metabolism, one of the most important metabolizing enzyme systems is cytochrome P450 (CYP) [36]. The results show that all compounds were substrates of CYP2C19, CYP2C9, and CYP2B6 isoforms, which may cause drug–drug interactions. Finally, regarding excretion, the half-lives (T1/2) of the compounds were used to assess their excretion rates. Drugs with T1/2 values close to 1 were characterized by a long half-life (>3 h), and T1/2 values close to 0 meant a shorter half-life (<3 h). The results reveal the tendency for half-life value reduction in the following order: D4 > D2 > D1 > D3. The combined ADMET results show that compounds D1, D2, and D4 fully comply with Lipinski’s [37], Ghose’s [38], Veber’s [39], Egan’s [40], and Muegge’s [41] rules, while compound D3 exceeds the molecular weight limit due to its naphthalene extension. However, its other properties (e.g., low LogP, acceptable TPSA) suggest potential bioavailability.

2.5. Molecular Docking

In order to investigate the difference in binding affinity and binding modes in various ligand–D3R complexes, interaction analyses were performed for the molecular docking results of D1, D2, D3, and D4 with D3R (see Figure 7). Compared with the template compound 30r, all designed compounds were scored better for docking with D3R than D2R (see Table S4), except compound D2, indicating that all compounds could bind to D3R more stably and form stronger binding interactions.
As shown in Figure 7, the docking results demonstrate that all the designed compounds exhibited stable binding within the active pockets of D3R, establishing various interactions with key residues. Notably, the left PCPMA motif of compounds D1, D3, and D4 was located at the orthosteric binding sites (OBS), which were characterized by hydrophobic pockets and surrounded by aromatic residues (PHE188, PHE345, PHE346, HIS349, and TYR365). At the same time, the PCPMA motif formed a stable hydrophobic interaction by π-π stacking with PHE345 and PHE346. And the protonated nitrogen atom of the linker chain formed a salt bridge or hydrogen bonds with the carboxylate of ASP110, offering a critical strong interaction, which was manifested in D1, D2, D3, and D4. Furthermore, the right aryl substituent was oriented toward the secondary binding pocket (SBP), which was constructed by extracellular loop1, loop2, and helices I, II, and VII; the nitrogen atoms of compound D3 in the amide bond of the linker formed hydrogen bonds. Moreover, in compound D4, hydrogen bonds and π-π stacking were formed between the nitrogen atom and the phenyl of the benzimidazole group with TYR365, which further stabilized the binding conformation, while the template compound 30r and compound D2 reversed the typical binding conformation at the OBS and SBP, which may be related to the biological activity of the compounds [17]. (The binding model of compound 30r with D3R can be seen in Figure S1.) On the other hand, there were basic interactions of hydrogen bonds and salt bridges between the above two compounds and the residues, ensuring the affinity of the compounds with D3R (see Figure 7 and Figure S1). Overall, the docking poses of all compounds showed strong binding to D3R, highlighting the importance of specific interactions in the D3R–PCPMA ligands complex that drove their biological efficacy.
This suggests that the designed compounds have enhanced binding capabilities, allowing them to interact more effectively and stably with the protein. The notable improvement in binding interactions implies that these compounds may form stronger interactions within the active site, which could lead to improved biological efficacy. Further analysis of these interactions would provide insights into the specific binding modes and contribute to the rational design of more potent ligands.

2.6. Molecular Dynamics (MD) Simulation

To verify the binding modes under dynamic conditions, 300 ns MD simulations were conducted for the best docking pose of the template compound 30r and the newly designed compounds D1, D2, D3, and D4 with D3R. The root mean square deviation (RMSD), the radius of gyration (Rg), the root mean square fluctuation (RMSF), and the solvent-accessible surface area (SASA) of the MD results were also analyzed. As shown in Figure 8, all designed compounds reached a stable binding state with the protein complex system after 125 ns of simulation, and the RMSD values fluctuated steadily between 0.25 and 0.3 nm. The D3R-D1 complex maintained a lower standard deviation (the RMSD for compound D1 was 0.282 ± 0.022; for D2, 0.283 ± 0.0181; for D3, 0.303 ± 0.0225; and for D4, 0.321 ± 0.026). Compared with the template compound 30r (see Figure S2), the newly designed compounds exhibited greater dynamic stability. Furthermore, during the whole simulation process, the RMSD of the D3R-D1 complex system had the lowest fluctuation and was more stable, indicating that the ligand–protein binding ability was stronger and the composite system was more stable than the other three systems, which is consistent with the docking results. These findings collectively indicate that the newly designed compounds exhibited a high affinity for D3R, establishing them as promising candidates for further investigation in the development of drugs targeting this receptor.
The RMSF values for the four systems are shown in Figure S3A: all systems exhibited the same trend, with the RMSF fluctuation trend being lower than that of the template compound 30r. The Rg characterizes the tightness of a protein structure and the change in the degree of peptide chain looseness of the protein during the simulation process. The results are shown in Figure S3B: with the exception of compound D4, the Rg fluctuation trends were more stable than those of the template compounds, suggesting that the four compounds bind to D3R with less overall change in the protein and less probability of folding and unfolding phenomena occurring. In addition, the SASA of the protein–ligand complexes were analyzed from the simulation trajectories to evaluate changes in the protein during simulation via each complex. As shown in Figure S3C, the SASA fluctuation trend was similar to that of the template compounds, except for compound D4, suggesting that the dynamic conformational changes of the protein expanded and contracted with the same trend throughout the simulations. Such characteristics were crucial for understanding the binding interactions and the potential functional implications of these compounds as therapeutic agents targeting D3R.

2.7. MMGB/PBSA Calculation of Binding Free Energy

To further measure the strength of interactions between the PCPMA ligands and D3R, MM/PBSA calculations were performed to explore the binding affinities for each ligand–D3R complex system. The last 25 ns of the MD simulation results were selected as the steady state to calculate the binding free energy, with the results shown in Table 4 ΔGvdW and ΔGpolar(PB) contributed to attractive interactions, while ∆Gsol was repulsive to the total binding free energies.
The ΔGtotal values of compounds D1 and D2 were lower than that of the template compound 30r, indicating more stable binding and stronger interaction with D3R. In contrast, the D3 and D4 systems had relatively high binding free energy, suggesting that these compounds may not bind as effectively as the template compound 30r to the protein. Notably, it could be seen that ΔGvdW was identified as important for ligand–receptor binding. In addition, ΔGpolar was detrimental to the binding affinity of the complexes because the large volume of the ligand binding pocket was exposed to a large amount of solvent.
The compound 30r PCPMA backbone is more exposed to solvents, resulting in higher polar solvation energy. The nitrogen atom of the PCPMA in compound D1 formed hydrogen bonds and salt bridges with ASP110, which enhanced its binding affinity with the proteins. This interaction led to a relatively smaller solvation energy and, therefore, a low total binding free energy. In compound D2, urea was added to the aliphatic linkage chain, enabling two nitrogen atoms to form hydrogen bonds and salt bridges with ASP110, and oxygen atoms formed hydrogen bonds with ILE183, which enhanced the binding interaction. At the same time, the molecule exhibited a strong electrostatic interaction, resulting in a larger ΔGgas value. However, due to the PCPMA backbone being more exposed to solvents, a larger ΔGpolar(PB) was obtained, leading to a relatively higher binding free energy. Compound D3 incorporated a large 2-(naphthale-2-yl)-ethyl group into the side chain of nitrogen atoms in the PCPMA framework, which allowed the molecule to better occupy the binding pocket. Furthermore, the introduction of triazole groups enabled the nitrogen atoms to form hydrogen bonds with ASN352, which enhanced the binding force, resulting in a higher ΔGvdW. In compound D4, the nitrogen atom of the imidazole group formed a hydrogen bond with TYR365, developing a strong electrostatic interaction, whereas a significant portion of the molecule was exposed to the solvent, leading to a larger ΔGpolar(PB).
To further identify the key residues of the D3R–PCPMA ligand interactions, the binding energies obtained from the MM–PBSA analysis were further decomposed into the contributions of per residue. The key residues that contributed to the binding free energy for the D3R compounds D1, D2, D3, and D4 complexes are shown in Figure 9. The analysis reveals that several residues in the vicinity of the binding site engaged in strong interactions with the designed compounds. Notably, the following residues were identified as having high energetic contributions to the binding free energy, thus playing critical roles in stabilizing the complexes: TYR26, LEU89, GLY94, VAL95, VAL98, VAL107, ASP110, VAL111, SER182, ILE183, PHE188, VAL189, SER192, SER196, PHE345, PHE346, HIS349, VAL360, PRO362, TYR365, SER366, THR369, and TYR373. According to the binding free energy decomposition results, the amino acid residues in the docking pocket contributed greatly to the binding free energy, and ASP110, which determines the affinity, was more prominent. These residues collectively demonstrated significant energetic contributions, indicating their involvement in various interactions that stabilize the ligand–receptor binding.

3. Materials and Methods

3.1. 3D-QSAR Analysis Dataset

A total of 50 different PCPMA derivatives were selected in this work, all of which were PCPMA derivatives reported by Cheng’s group [23]. Compound 30r, with the best affinity, was selected as the template molecule, and Gaussian 09 was used to optimize the structure at the level of B3LYP/6-31G(d) [42,43]. The other 49 molecules were constructed by changing the moieties of 30r and partially structurally optimized with Gaussian 09 (in which the common skeleton was fixed). All molecules were categorized into different classes, and one molecule at a time from each different type was selected for the test set. At the same time, the compounds were divided into a training set (40 compounds, 80%) and a test set (10 compounds, 20%), ensuring that both sets were structurally diverse with high, medium, and low activity. The distributions of the pKi values of the training and test sets are shown in Table S1.

3.2. Compound Preparation and Restructuring

After constructing the 3D structures of the molecules, energy minimization was carried out using the Sybyl-X 2.0 (SYBYL-X 2.0, Tripos International, St. Louis, MO, USA) software package with the Tripos force field and Gasteiger–Huckel charges and 0.005 kcal/mol gradient convergence conditions using the Powell conjugate gradient method with 1000 iterations. A common backbone-based ligand stacking method was chosen for backbone stacking [44,45].

3.3. 3D-QSAR Models Production

Stacked training set molecules were placed in a 3D cubic lattice with a grid spacing of 2.0 Å to generate CoMFA and CoMSIA descriptor fields. The CoMFA fields were generated using sp3 carbon probe atoms with a charge of +1.0, and a van der Waals radius of 1.52 Å at each grid point, calculated using Tripos, hydrogen bonding, and indicator force fields, using the Lennard–Jones potential term to calculate the space field and a Coulomb term to calculate the electrostatic field. The cutoff values for both the spatial and electrostatic fields were set to 30.0 kcal/mol, and the rest of the parameters were according to the reference values. The CoMSIA fields (with the probe charge of +1.0, radius of 1 Å, hydrophobicity of +1.0, and acceptor of hydrogen bonds of +1.0) were calculated using the standard settings to obtain the steric, electrostatic, hydrophobic, hydrogen bond acceptor, and hydrogen bond donor fields. The internal predictive ability of the CoMFA and CoMSIA models was performed on the training set using partial least squares (PLS) analysis. Cross-validation correlation coefficients (q2) were determined by a leave-one-out (LOO) cross-validation procedure, and correlation coefficients (r2) were calculated by non-cross-validation methods. All CoMFA and CoMSIA studies were conducted using Sybyl-X 2.0 software.

3.4. Design of Novel PCPMA-Like D3R-Selective Ligands and Their Affinity Prediction with D3R

After constructing and validating the 3D-QSAR model, the relationship between molecular structure and binding affinity with D3R was illustrated based on the contour map information provided by the above model. Furthermore, a series of novel PCPMA-like D3R-selective ligands could be designed and their binding affinity activity with D3R could be predicted. The process was completed by Sybyl-X 2.0 and the newly designed ligands were treated as per Section 2.2.

3.5. Pharmacokinetic ADMET and Drug Similarity Prediction

The designed PCPMA derivatives were further investigated using ADMET prediction to assess the properties such as bioavailability, solubility, plasma protein binding, etc. Several pharmacokinetic parameters such as absorption, skin penetration, blood–brain barrier penetration, cytochrome P450 (CYP), toxicity, synthetic accessibility, and drug similarity prediction were determined using ADMETlab 3.0 [34] (https://admetlab3.scbdd.com/, accessed on 10 March 2025).

3.6. Molecular Docking

Molecular docking was performed using the Schrodinger 2021-2 software. Firstly, the 3D structures of the compounds were established. Then the 3D structures of these compounds were energy minimized by Gaussian 09. The crystal structure of the D3R target protein for docking was obtained from the PDB database (http://www.rcsb.org, accessed on 1 March 2023); the PDB ID is 3PBL [4]. The D3R structure was a dimer of the same structure, the intracellular structural domain was excised, and all remaining structural elements were identical to the crystal structure. Since the aim of our study was to investigate the effects of different ligands on the receptor, using the same variations in the receptor still did not bias the results, and the fact of the ligand binding to the transmembrane structural domains instead of the intracellular structural domains had no effect on the computation results, so the A chain of the subsequent D3R receptor was used for docking calculations. Afterward, the D3R protein structure was prepared by the Protein Preparation Wizard module in the Schrodinger 2021-2 software. The protein was assigned bond orders, had hydrogens added, was protonated, and had all crystallographic water molecules removed, and restrained minimization was applied. The compounds were structurally optimized using Gaussian 09 at the B3LYP/6-31G(d) level before docking and then were prepared using LigPrep with default settings. The docking grid was prepared with Glide [46], defining the binding site by crystal ligands and setting the ligand diameter midpoint box to 20 Å on all three axes. Finally, all of these ligands were docked into the calculated receptor grid using the XP scoring function and enhanced sampling [47]. The docking results were visualized and analyzed in Maestro, and the best scoring poses were selected for further dynamic simulation analysis.

3.7. Molecular Dynamics Simulations (MD)

The ligand–D3R complexes with the lowest absolute binding free energy in molecular docking were selected for molecular dynamics simulations using Gromacs 2023 [48], employing the frog hopping Newton integral algorithm for equilibrium kinetic integration. During the simulation, the intracellular structural domain of the D3R structure was excised and all remaining structural elements were identical to the crystal structure. Amber14sb_parmbsc1 force field parameters were used to generate protein topologies, and ligands were firstly calculated using Multiwfn [49] for RESP2(0.5) (http://sobereva.com/476, accessed on 1 May 2024) charges, and then the Sobtop (http://sobereva.com/soft/Sobtop, accessed on 1 May 2024) software was used to generate the GAFF force field topology. The TIP3P water model was used to add solvent to the protein–ligand system. Then, cubic boxes were built and replenished with Na+/Cl− to balance the system. The energy of the complex system was optimized using the 50,000-step most rapid descent method (SD) combined with conjugate gradient (CG) methods. After the system energy optimization was completed, 100 ps limiting dynamics simulations (including isovolumetric NPT simulations) were performed for the composite system at one atmosphere and a temperature of 298.15 °C. Finally, a 300 ns molecular dynamics simulation was performed (the simulation was performed with a time step of 2 fs and the simulation trajectory was saved every 50 ps), and the trajectories were used to analyze the root mean square deviation (RMSD), root mean square fluctuation (RMSF), solvent accessible surface area (SASA), radius of gyration (Rg), etc.

3.8. Combined Free Energy MMGB/PBSA Calculations and Per-Residue Free Energy Decomposition Analysis

The molecular mechanics/Poisson–Boltzmann surface area (MM-PBSA) [50] calculations were preformed to estimate the ligand–protein binding free energy after MD simulations using the gmx_MMPBSA [51] module of the AmberTools 20 software. Based on the trajectory file of the MD simulation, the last 25 ns of the stable trajectory was selected for binding energy calculation, and 1 frame was selected for every 100 ps, for a total of 250 frames. The binding free energy calculations (ΔGtotal) focused on the molecular mechanical energy (ΔEmm) and solvation free energy (ΔGsol), and entropy calculations (TΔS) for the binding free energy calculations were taken from previously published studies [52,53]. The molecular mechanical energy was further decomposed into van der Waals energy (ΔEvdW), electrostatic energy (ΔEele) and solvation free energy (ΔGsol), the last of which was decomposed into the contribution of polar solvation energy and non-polar solvation energy. In order to estimate the decomposition parameters for the van der Waals (ΔGvdW), electrostatic (ΔGele), polar (ΔGpolar), and nonpolar (ΔGnonpolar) energies, the contribution of amino acid residues interacting with the receptor to binding free energy was studied.

4. Conclusions

As a promising target for the treatment of psychiatric diseases, research on the dopamine D3R receptor has primarily focused on the development of ligands with high selectivity and high affinity. Low selectivity would bring a series of adverse side effects, and low affinity would lead to the reduction of ligand efficacy; therefore, an in-depth computational study was conducted on the selective dopamine D3R ligands. In this study, predictive 3D-QSAR models were developed using CoMFA and CoMSIA methods with 50 selective D3R ligands to identify and design novel PCPMA derivatives as D3R ligands. Both CoMFA and CoMSIA models yielded satisfactory predictive outcomes, with q2 of 0.607 and 0.643, r2 of 0.981 and 0.899, respectively, evaluated by a series of statistical parameters. The results of the comparison between the CoMFA and CoMSIA models show that CoMFA was superior to the CoMSIA model, indicating that the electrostatic and steric fields had an important impact on the affinity between D3R and PCPMA ligands. The contour plots provided further information to illustrate the structure–activity relationships: when the side chains of the nitrogen atoms in the connecting chain were replaced by naphthalene, it was predicted to have higher affinity. The affinity of the compound increased when the group attached after the amide bond was a hydrophilic, negatively charged triazole and imidazole group, a finding which will assist in the rational design of new PCPMA derivatives. Additionally, based on the 3D-QSAR modeling results, four novel D3R ligands were designed, and the prediction results show that their affinity with D3R was improved. To further validate the above predictions, molecular docking and 300 ns molecular dynamics simulations were conducted to evaluate the stability of the D3R–ligand complexes. The MD results show low RMSD and RMSF values, which indicate conformational stability throughout the simulation, and the above stability was in good agreement with the 3D-QSAR model and the molecular docking results. Furthermore, the binding free energies calculated by the MM-PBSA method further confirmed that the binding affinities of the designed compounds D1 and D2 to D3R were higher than that of the template molecule.
Overall, this work provides useful guidance for the rational design of selective D3R ligands and better explains the complex structure–activity relationships that exist between these selective D3R ligands, which can then be used to design novel, more potent, and more stable D3R ligands, and provides a theoretical basis for the discovery of candidate molecules with potential antipsychotic activity to treat patients with a variety of diseases, such as Parkinson’s disease or schizophrenia.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/ijms26083559/s1. Refs. [54,55] are cited in Supplementary Materials.

Author Contributions

Conceptualization, W.Z. and X.P; methodology, L.G. and W.Z.; software, L.G. and Y.G.; validation, L.G., W.Z. and S.Z.; formal analysis, R.Z.; investigation, L.Z.; resources, W.Z. and X.P.; data curation, L.G. and R.Z.; writing-original draft preparation, L.G.; writing-review and editing, W.Z.; visualization, X.P. and P.S.; supervision, P.S., W.Z. and X.P.; project administration, W.Z.; funding acquisition, W.Z. All authors have read and agreed to the published version of the manuscript.

Funding

This research was supported by the Shihezi University High-level Talent Research Start-up funds (2022ZK021), and the Tianchi Talent Introduction Plan—Young Doctor Talents (CZ000902, CZ000917).

Informed Consent Statement

Not applicable.

Data Availability Statement

The data presented in this study are available in the article.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. Levant, B. The D3 dopamine receptor: Neurobiology and potential clinical relevance. Pharmacol. Rev. 1997, 49, 231–252. [Google Scholar] [CrossRef] [PubMed]
  2. Löber, S.; Hübner, H.; Tschammer, N.; Gmeiner, P. Recent advances in the search for D3- and D4-selective drugs: Probes, models and candidates. Trends Pharmacol. Sci. 2011, 32, 148–157. [Google Scholar] [CrossRef]
  3. Sokoloff, P.; Giros, B.; Martres, M.-P.; Bouthenet, M.-L.; Schwartz, J.-C. Molecular cloning and characterization of a novel dopamine receptor (D3) as a target for neuroleptics. Nature 1990, 347, 146–151. [Google Scholar] [CrossRef]
  4. Chien, E.Y.T.; Liu, W.; Zhao, Q.; Katritch, V.; Han, G.W.; Hanson, M.A.; Shi, L.; Newman, A.H.; Javitch, J.A.; Cherezov, V.; et al. Structure of the human dopamine D3 receptor in complex with a D2/D3 selective antagonist. Science 2010, 330, 1091–1095. [Google Scholar] [CrossRef]
  5. Sokoloff, P.; Giros, B.; Martres, M.P.; Andrieux, M.; Besancon, R.; Pilon, C.; Bouthenet, M.L.; Souil, E.; Schwartz, J.C. Localization and function of the D3 dopamine receptor. Arzneimittelforschung 1992, 42, 224–230. [Google Scholar]
  6. Zingales, V.; Torrisi, S.A.; Leggio, G.M.; Bucolo, C.; Drago, F.; Salomone, S. Pharmacological and Genetic Evidence of Dopamine Receptor 3-Mediated Vasoconstriction in Isolated Mouse Aorta. Biomolecules 2021, 11, 418. [Google Scholar] [CrossRef]
  7. Reilly, S.W.; Riad, A.A.; Hsieh, C.-J.; Sahlholm, K.; Jacome, D.A.; Griffin, S.; Taylor, M.; Weng, C.-C.; Xu, K.; Kirschner, N.; et al. Leveraging a Low-Affinity Diazaspiro Orthosteric Fragment to Reduce Dopamine D3 Receptor (D3R) Ligand Promiscuity across Highly Conserved Aminergic G-Protein-Coupled Receptors (GPCRs). J. Med. Chem. 2019, 62, 5132–5147. [Google Scholar] [CrossRef]
  8. Keck, T.M.; John, W.S.; Czoty, P.W.; Nader, M.A.; Newman, A.H. Identifying Medication Targets for Psychostimulant Addiction: Unraveling the Dopamine D3 Receptor Hypothesis. J. Med. Chem. 2015, 58, 5361–5380. [Google Scholar] [CrossRef]
  9. Stemp, G.; Ashmeade, T.; Branch, C.L.; Hadley, M.S.; Hunter, A.J.; Johnson, C.N.; Nash, D.J.; Thewlis, K.M.; Vong, A.K.K.; Austin, N.E.; et al. Design and synthesis of trans-N-4-2-(6-cyano-1,2,3,4-tetrahydroisoquinolin-2-yl) ethyl cyclohexyl -4-quinolinecarboxamide (SB-277011): A potent and selective dopamine D3 receptor antagonist with high oral bioavailability and CNS penetration in the rat. J. Med. Chem. 2000, 43, 1878–1885. [Google Scholar] [CrossRef]
  10. Appel, N.M.; Li, S.-H.; Holmes, T.H.; Acri, J.B. Dopamine D3 Receptor Antagonist (GSK598809) Potentiates the Hypertensive Effects of Cocaine in Conscious, Freely-Moving Dogs. J. Pharmacol. Exp. Ther. 2015, 354, 484–492. [Google Scholar] [CrossRef]
  11. Nielsen, S.; Gowing, L.; Sabioni, P.; Le Foll, B. Pharmacotherapies for cannabis dependence. Cochrane Database Syst. Rev. 2019, 1, CD008940. [Google Scholar] [CrossRef] [PubMed]
  12. Gyertyán, I.; Kiss, B.; Sághy, K.; Laszy, J.; Szabó, G.; Szabados, T.; Gémesi, L.I.; Pásztor, G.; Zájer-Balázs, M.; Kapás, M.; et al. Cariprazine (RGH-188), a potent D3/D2 dopamine receptor partial agonist, binds to dopamine D3 receptors in vivo and shows antipsychotic-like and procognitive effects in rodents. Neurochem. Int. 2011, 59, 925–935. [Google Scholar] [CrossRef] [PubMed]
  13. Shonberg, J.; Lopez, L.; Scammells, P.J.; Christopoulos, A.; Capuano, B.; Lane, J.R. Biased agonism at G protein-coupled receptors: The promise and the challenges—A medicinal chemistry perspective. Med. Res. Rev. 2014, 34, 1286–1330. [Google Scholar] [CrossRef] [PubMed]
  14. Lane, J.R.; Sexton, P.M.; Christopoulos, A. Bridging the gap: Bitopic ligands of G-protein-coupled receptors. Trends Pharmacol. Sci. 2013, 34, 59–66. [Google Scholar] [CrossRef]
  15. Cao, Y.; Sun, N.; Zhang, J.; Liu, Z.; Tang, Y.-Z.; Wu, Z.; Kim, K.-M.; Cheon, S.H. Design, synthesis, and evaluation of bitopic arylpiperazine-phthalimides as selective dopamine D3 receptor agonists. Medchemcomm 2018, 9, 1457–1465. [Google Scholar] [CrossRef]
  16. Shaik, A.B.; Boateng, C.A.; Battiti, F.O.; Bonifazi, A.; Cao, J.; Chen, L.; Chitsazi, R.; Ravi, S.; Lee, K.H.; Shi, L.; et al. Structure Activity Relationships for a Series of Eticlopride-Based Dopamine D2/D3 Receptor Bitopic Ligands. J. Med. Chem. 2021, 64, 15313–15333. [Google Scholar] [CrossRef]
  17. Arroyo-Urea, S.; Nazarova, A.L.; Carrión-Antolí, Á.; Bonifazi, A.; Battiti, F.O.; Lam, J.H.; Newman, A.H.; Katritch, V.; García-Nafría, J. A bitopic agonist bound to the dopamine 3 receptor reveals a selectivity site. Nat. Commun. 2024, 15, 7759. [Google Scholar] [CrossRef]
  18. Dalton King, H.; Denhart, D.J.; Deskus, J.A.; Ditta, J.L.; Epperson, J.R.; Higgins, M.A.; Kung, J.E.; Marcin, L.R.; Sloan, C.P.; Mattson, G.K.; et al. Conformationally restricted homotryptamines. Part 4: Heterocyclic and naphthyl analogs of a potent selective serotonin reuptake inhibitor. Bioorg. Med. Chem. Lett. 2007, 17, 5647–5651. [Google Scholar] [CrossRef]
  19. Liu, J.; Clough, S.J.; Hutchinson, A.J.; Adamah-Biassi, E.B.; Popovska-Gorevski, M.; Dubocovich, M.L. MT1 and MT2 Melatonin Receptors: A Therapeutic Perspective. Annu. Rev. Pharmacool. Toxicol. 2016, 56, 361–383. [Google Scholar] [CrossRef]
  20. Cho, S.J.; Jensen, N.H.; Kurome, T.; Kadari, S.; Manzano, M.L.; Malberg, J.E.; Caldarone, B.; Roth, B.L.; Kozikowski, A.P. Selective 5-Hydroxytryptamine 2C Receptor Agonists Derived from the Lead Compound Tranylcypromine: Identification of Drugs with Antidepressant-Like Action. J. Med. Chem. 2009, 52, 1885–1902. [Google Scholar] [CrossRef]
  21. Carpenter, J.; Wang, Y.; Wu, G.; Feng, J.; Ye, X.-Y.; Morales, C.L.; Broekema, M.; Rossi, K.A.; Miller, K.J.; Murphy, B.J.; et al. Utilization of an Active Site Mutant Receptor for the Identification of Potent and Selective Atypical 5-HT 2C Receptor Agonists. J. Med. Chem. 2017, 60, 6166–6190. [Google Scholar] [CrossRef] [PubMed]
  22. Zhang, G.; Cheng, J.; McCorvy, J.D.; Lorello, P.J.; Caldarone, B.J.; Roth, B.L.; Kozikowski, A.P. Discovery of N-Substituted (2-Phenylcyclopropyl)methylamines as Functionally Selective Serotonin 2C Receptor Agonists for Potential Use as Antipsychotic Medications. J. Med. Chem. 2017, 60, 6273–6288. [Google Scholar] [CrossRef] [PubMed]
  23. Tan, L.; Zhou, Q.; Yan, W.; Sun, J.; Kozikowski, A.P.; Zhao, S.; Huang, X.-P.; Cheng, J. Design and Synthesis of Bitopic 2-Phenylcyclopropylmethylamine (PCPMA) Derivatives as Selective Dopamine D3 Receptor Ligands. J. Med. Chem. 2020, 63, 4579–4602. [Google Scholar] [CrossRef] [PubMed]
  24. Zhao, L.; Zhang, L.; Lei, M. 3D-QSAR and docking studies on 2-arylbenzoxazole and linker-Y transthyretin amyloidogenesis inhibitors. Sci. China Chem. 2013, 56, 1550–1563. [Google Scholar] [CrossRef]
  25. Damale, M.G.; Harke, S.N.; Kalam Khan, F.A.; Shinde, D.B.; Sangshetti, J.N. Recent advances in multidimensional QSAR (4D-6D): A critical review. Mini Rev. Med. Chem. 2014, 14, 35–55. [Google Scholar] [CrossRef]
  26. Lambrinidis, G.; Vallianatou, T.; Tsantili-Kakoulidou, A. In vitro, in silico and integrated strategies for the estimation of plasma protein binding. A review. Adv. Drug Deliv. Rev. 2015, 86, 27–45. [Google Scholar] [CrossRef]
  27. Kim, K.H. Comparative molecular field analysis (CoMFA). In Molecular Similarity in Drug Design; Springer: Dordrecht, The Netherlands, 1995; pp. 291–331. [Google Scholar]
  28. Klebe, G.; Abraham, U.; Mietzner, T. Molecular similarity indices in a comparative analysis (CoMSIA) of drug molecules to correlate and predict their biological activity. J. Med. Chem. 1994, 37, 4130–4146. [Google Scholar] [CrossRef]
  29. Völkner, M.; Wagner, F.; Steinheuer, L.M.; Carido, M.; Kurth, T.; Yazbeck, A.; Schor, J.; Wieneke, S.; Ebner, L.J.A.; Del Toro Runzer, C.; et al. HBEGF-TNF induce a complex outer retinal pathology with photoreceptor cell extrusion in human organoids. Nat. Commun. 2022, 13, 6183. [Google Scholar] [CrossRef]
  30. Katritch, V.; Cherezov, V.; Stevens, R.C. Structure-Function of the G Protein–Coupled Receptor Superfamily. Annu. Rev. Pharmacol. Toxicol. 2013, 53, 531–556. [Google Scholar] [CrossRef]
  31. Chen, L.; Lin, Y.; Yan, X.; Ni, H.; Chen, F.; He, F. 3D-QSAR studies on the structure–bitterness analysis of citrus flavonoids. Food Funct. 2023, 14, 4921–4930. [Google Scholar] [CrossRef]
  32. Salas, C.O.; Zarate, A.M.; Kryštof, V.; Mella, J.; Faundez, M.; Brea, J.; Loza, M.I.; Brito, I.; Hendrychová, D.; Jorda, R.; et al. Promising 2,6,9-Trisubstituted Purine Derivatives for Anticancer Compounds: Synthesis, 3D-QSAR, and Preliminary Biological Assays. Int. J. Mol. Sci. 2019, 21, 161. [Google Scholar] [CrossRef] [PubMed]
  33. Zięba, A.; Laitinen, T.; Patel, J.Z.; Poso, A.; Kaczor, A.A. Docking-Based 3D-QSAR Studies for 1,3,4-oxadiazol-2-one Derivatives as FAAH Inhibitors. Int. J. Mol. Sci. 2021, 22, 6108. [Google Scholar] [CrossRef] [PubMed]
  34. Fu, L.; Shi, S.; Yi, J.; Wang, N.; He, Y.; Wu, Z.; Peng, J.; Deng, Y.; Wang, W.; Wu, C.; et al. ADMETlab 3.0: An updated comprehensive online ADMET prediction platform enhanced with broader coverage, improved performance, API functionality and decision support. Nucleic Acids Res. 2024, 52, W422–W431. [Google Scholar] [CrossRef]
  35. Smith, D.A.; Beaumont, K.; Maurer, T.S.; Di, L. Volume of Distribution in Drug Design. J. Med. Chem. 2015, 58, 5691–5698. [Google Scholar] [CrossRef]
  36. Hollenberg, P.F. Mechanisms of cytochrome P450 and peroxidase-catalyzed xenobiotic metabolism. FASEB J. 1992, 6, 686–694. [Google Scholar] [CrossRef]
  37. Lipinski, C. Interview by Peter Kirkpatrick. Nat. Rev. Drug Discov. 2012, 11, 900–901. [Google Scholar] [CrossRef]
  38. Ghose, A.K.; Viswanadhan, V.N.; Wendoloski, J.J. A Knowledge-Based Approach in Designing Combinatorial or Medicinal Chemistry Libraries for Drug Discovery. 1. A Qualitative and Quantitative Characterization of Known Drug Databases. J. Comb. Chem. 1999, 1, 55–68. [Google Scholar] [CrossRef]
  39. Plinski, E.F.; Plinska, S. Veber’s Rules in Terahertz Light. 2020. Available online: https://www.researchsquare.com/article/rs-12857/v1 (accessed on 4 April 2024).
  40. Egan, W.J.; Merz, K.M.; Baldwin, J.J. Prediction of Drug Absorption Using Multivariate Statistics. J. Med. Chem. 2000, 43, 3867–3877. [Google Scholar] [CrossRef]
  41. Muegge, I. Selection criteria for drug-like compounds. Med. Res. Rev. 2003, 23, 302–321. [Google Scholar] [CrossRef]
  42. Becke, A.D. Density-functional thermochemistry. III. The role of exact exchange. J. Chem. Phys. 1993, 98, 5648–5652. [Google Scholar] [CrossRef]
  43. Hariharan, P.C.; Pople, J.A. The influence of polarization functions on molecular orbital hydrogenation energies. Theor. Chim. Acta 1973, 28, 213–222. [Google Scholar] [CrossRef]
  44. Lanka, G.; Begum, D.; Banerjee, S.; Adhikari, N.; P, Y.; Ghosh, B. Pharmacophore-based virtual screening, 3D QSAR, Docking, ADMET, and MD simulation studies: An in silico perspective for the identification of new potential HDAC3 inhibitors. Comput. Biol. Med. 2023, 166, 107481. [Google Scholar] [CrossRef] [PubMed]
  45. Yan, W.; Lin, G.; Zhang, R.; Liang, Z.; Wu, L.; Wu, W. Studies on molecular mechanism between ACE and inhibitory peptides in different bioactivities by 3D-QSAR and MD simulations. J. Mol. Liq. 2020, 304, 112702. [Google Scholar] [CrossRef]
  46. Repasky, M.P.; Shelley, M.; Friesner, R.A. Flexible Ligand Docking with Glide. Curr. Protoc. Bioinform. 2007, 18, 8–12. [Google Scholar] [CrossRef]
  47. Friesner, R.A.; Murphy, R.B.; Repasky, M.P.; Frye, L.L.; Greenwood, J.R.; Halgren, T.A.; Sanschagrin, P.C.; Mainz, D.T. Extra Precision Glide: Docking and Scoring Incorporating a Model of Hydrophobic Enclosure for Protein-Ligand Complexes. J. Med. Chem. 2006, 49, 6177–6196. [Google Scholar] [CrossRef]
  48. Van Der Spoel, D.; Lindahl, E.; Hess, B.; Groenhof, G.; Mark, A.E.; Berendsen, H.J.C. GROMACS: Fast, flexible, and free. J. Comput. Chem. 2005, 26, 1701–1718. [Google Scholar] [CrossRef]
  49. Lu, T.; Chen, F. Multiwfn: A multifunctional wavefunction analyzer. J. Comput. Chem. 2012, 33, 580–592. [Google Scholar] [CrossRef]
  50. Chen, J.; Zeng, Q.; Wang, W.; Sun, H.; Hu, G. Decoding the Identification Mechanism of an SAM-III Riboswitch on Ligands through Multiple Independent Gaussian-Accelerated Molecular Dynamics Simulations. J. Chem. Inf. Model. 2022, 62, 6118–6132. [Google Scholar] [CrossRef]
  51. Valdés-Tresanco, M.S.; Valdés-Tresanco, M.E.; Valiente, P.A.; Moreno, E. gmx_MMPBSA: A New Tool to Perform End-State Free Energy Calculations w ith GROMACS. J. Chem. Theory Comput. 2021, 17, 6281–6291. [Google Scholar] [CrossRef]
  52. Chohan, T.A.; Chen, J.-J.; Qian, H.-Y.; Pan, Y.-L.; Chen, J.-Z. Molecular modeling studies to characterize N-phenylpyrimidin-2-amine selectivity for CDK2 and CDK4 through 3D-QSAR and molecular dynamics simulations. Mol. Biosyst. 2016, 12, 1250–1268. [Google Scholar] [CrossRef]
  53. Chohan, T.A.; Qian, H.-Y.; Pan, Y.-L.; Chen, J.-Z. Molecular simulation studies on the binding selectivity of 2-anilino-4 -(thiazol-5-yl)-pyrimidines in complexes with CDK2 and CDK7. Mol. Biosyst. 2016, 12, 145–161. [Google Scholar] [CrossRef] [PubMed]
  54. Cramer, R.D.; Patterson, D.E.; Bunce, J.D. Comparative molecular field analysis (CoMFA). 1. Effect of shape on binding of steroids to carrier proteins. J. Am. Chem. Soc. 1988, 110, 5959–5967. [Google Scholar] [CrossRef] [PubMed]
  55. Kubinyi, H. QSAR and 3D QSAR in drug design Part 1: Methodology. Drug Discovery Today 1997, 2, 457–467. [Google Scholar] [CrossRef]
Ijms 26 03559 g001
Figure 1. (A) D3R-selective ligands in clinic and clinical trials. (B) The bitopic ligands of D3R.
Figure 1. (A) D3R-selective ligands in clinic and clinical trials. (B) The bitopic ligands of D3R.
Ijms 26 03559 g001
Ijms 26 03559 sch001
Scheme 1. The workflow of the current study.
Scheme 1. The workflow of the current study.
Ijms 26 03559 sch001
Ijms 26 03559 g002
Figure 2. (A) The structure of compound 30r and structure-based aligned common substructures are shown in red bold lines. (B) Structural superposition of 50 ligands for 3D-QSAR modeling.
Figure 2. (A) The structure of compound 30r and structure-based aligned common substructures are shown in red bold lines. (B) Structural superposition of 50 ligands for 3D-QSAR modeling.
Ijms 26 03559 g002
Ijms 26 03559 g003
Figure 3. Scatter plots of experimental and predicted affinity (pKi) for (A) CoMFA and (B) COMSIA models.
Figure 3. Scatter plots of experimental and predicted affinity (pKi) for (A) CoMFA and (B) COMSIA models.
Ijms 26 03559 g003
Ijms 26 03559 g004
Figure 4. (A) CoMFA steric field and (B) electrostatic field contour map of 30r.
Figure 4. (A) CoMFA steric field and (B) electrostatic field contour map of 30r.
Ijms 26 03559 g004
Ijms 26 03559 g005
Figure 5. (A) CoMSIA steric field and (B) hydrophobic field contour map of 30r.
Figure 5. (A) CoMSIA steric field and (B) hydrophobic field contour map of 30r.
Ijms 26 03559 g005
Ijms 26 03559 g006
Figure 6. Structural modification sites of compound 30r.
Figure 6. Structural modification sites of compound 30r.
Ijms 26 03559 g006
Ijms 26 03559 g007
Figure 7. Predicted three-dimensional (3D) binding models of designed compounds (A) D1, (B) D2, (C) D3, and (D) D4 with D3R.
Figure 7. Predicted three-dimensional (3D) binding models of designed compounds (A) D1, (B) D2, (C) D3, and (D) D4 with D3R.
Ijms 26 03559 g007
Ijms 26 03559 g008
Figure 8. The RMSD of molecular dynamics simulations for the newly designed (A) compound D1, (B) compound D2, (C) compound D3, and (D) compound D4. Lig Fit Prot: Ligand-fitted protein.
Figure 8. The RMSD of molecular dynamics simulations for the newly designed (A) compound D1, (B) compound D2, (C) compound D3, and (D) compound D4. Lig Fit Prot: Ligand-fitted protein.
Ijms 26 03559 g008
Ijms 26 03559 g009
Figure 9. Binding free energy decomposition for the four protein–ligand complexes.
Figure 9. Binding free energy decomposition for the four protein–ligand complexes.
Ijms 26 03559 g009
Table 1. The statistical parameters of CoMFA and the best CoMSIA models.
Table 1. The statistical parameters of CoMFA and the best CoMSIA models.
PLS StatisticCoMFACoMSIA
q20.6070.643
r20.9810.889
SEE0.0940.207
F149.22269.912
N104
Steric0.3930.375
Electrostatic0.607-
Hydrophobic-0.725
SEP0.50.51
RMSE0.280.35
Note: q2: Non-cross-validated correlation coefficient, r2: Cross-validated correlation coefficient, SEE: Standard error of the estimate, F: F-test value, N: Optimum number of components, SEP: Standard error of prediction, RMSE: Root mean square error. When q2 > 0.5, r2 > 0.8, SEE ≈ 1, and F > 100, it indicates that the model is reliable.
Table 2. Designed ligands based on 3D-QSAR model and prediction results.
Table 2. Designed ligands based on 3D-QSAR model and prediction results.
CompoundsStructurespKi (Pre.)
CoMFACoMSIA
30rIjms 26 03559 i0018.8458.723
D1Ijms 26 03559 i0028.9668.706
D2Ijms 26 03559 i0038.8718.586
D3Ijms 26 03559 i0048.8698.248
D4Ijms 26 03559 i0058.9258.379
Table 3. Prediction of molecular property parameters of template molecule 30r and the newly designed compounds.
Table 3. Prediction of molecular property parameters of template molecule 30r and the newly designed compounds.
30rD1D2D3D4
Molecular weight509.2499.19514.29611.22472.18
nRot1313141412
Flexibility0.5910.6190.6670.4380.6
LogP6.3255.275.5256.7555.917
TPSA 1/Å245.2363.0575.0863.0561.02
SAscore 2<6<6<6<6<6
nHA44765
nHD11212
nRot: number of rotatable bonds, optimal is 0–11; Flexibility: nRot/nRig, number of rotatable bonds/number of rings; 1 TPSA: Topological polar surface area, 2 SAscore: Synthetic accessibility score (SAscore < 6, easy to synthesize); nHA: number of hydrogen bond acceptors; nHD: number of hydrogen bond donors.
Table 4. The binding free energies of compounds D1, D2, D3, and D4 to D3R and calculations for each energy term. All energies are expressed in kcal/mol.
Table 4. The binding free energies of compounds D1, D2, D3, and D4 to D3R and calculations for each energy term. All energies are expressed in kcal/mol.
System30rD1D2D3D4
ΔGvdW−72.88−69.53−62.28−70.08−58.04
ΔGele−16.68−20.65−46.98−14.39−37.35
ΔGgas−89.56−90.18−109.26−84.47−95.39
ΔGpolar(PB)51.8648.4169.1550.5361.19
ΔGnonpolar−6.28−5.58−6.27−6.63−5.56
ΔGsol45.5842.8362.8943.8955.64
ΔGtotal−43.98 −47.35−46.36−40.58−39.75
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Guo, L.; Gao, Y.; Zhang, S.; Zhao, L.; Zhao, R.; Sun, P.; Pan, X.; Zhang, W. 2-Phenylcyclopropylmethylamine (PCPMA) Derivatives as D3R-Selective Ligands for 3D-QSAR, Docking and Molecular Dynamics Simulation Studies. Int. J. Mol. Sci. 2025, 26, 3559. https://doi.org/10.3390/ijms26083559

AMA Style

Guo L, Gao Y, Zhang S, Zhao L, Zhao R, Sun P, Pan X, Zhang W. 2-Phenylcyclopropylmethylamine (PCPMA) Derivatives as D3R-Selective Ligands for 3D-QSAR, Docking and Molecular Dynamics Simulation Studies. International Journal of Molecular Sciences. 2025; 26(8):3559. https://doi.org/10.3390/ijms26083559

Chicago/Turabian Style

Guo, Li, Yuepeng Gao, Sujuan Zhang, Lingmi Zhao, Runxin Zhao, Pinghua Sun, Xinhui Pan, and Wei Zhang. 2025. "2-Phenylcyclopropylmethylamine (PCPMA) Derivatives as D3R-Selective Ligands for 3D-QSAR, Docking and Molecular Dynamics Simulation Studies" International Journal of Molecular Sciences 26, no. 8: 3559. https://doi.org/10.3390/ijms26083559

APA Style

Guo, L., Gao, Y., Zhang, S., Zhao, L., Zhao, R., Sun, P., Pan, X., & Zhang, W. (2025). 2-Phenylcyclopropylmethylamine (PCPMA) Derivatives as D3R-Selective Ligands for 3D-QSAR, Docking and Molecular Dynamics Simulation Studies. International Journal of Molecular Sciences, 26(8), 3559. https://doi.org/10.3390/ijms26083559

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop