Identification of potent allosteric inhibitors of PRMT3: pharmacophore based 3D-QSAR modeling and molecular simulations approach

doi:10.21203/rs.3.rs-3556197/v1

Download PDF

Research Article

Identification of potent allosteric inhibitors of PRMT3: pharmacophore based 3D-QSAR modeling and molecular simulations approach

https://doi.org/10.21203/rs.3.rs-3556197/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Protein arginine methyltransferases are a family of proteins that play an important role in the regulation of gene expression, mainly involved in epigenetic modifications like DNA methylation and histone acetylation. Protein arginine methyltransferase 3 (PRMT) is a member of the PRMT family of enzymes involved in various cellular processes, including transcriptional regulation, RNA processing, and signal transduction. PRMT3 levels are considerably enhanced in certain types of cancers particularly breast, colorectal, lung and ovarian cancers. Studies have indicated that PRMT3 promotes tumor growth by activating oncogenes and suppressing tumor suppressor genes. PRMT3 is a histone-modifying enzyme that catalyzes the transfer of methyl groups from S-adenosylmethionine (SAM) to specific arginine residues in target proteins, thereby modulating their function. However, the activity of PRMT3 is not only regulated by its catalytic domain but also by allosteric mechanisms that control its enzymatic activity. Furthermore, targeting PRMT3 could potentially serve as a therapeutic strategy for treating various types of cancer. This study focuses on targeting the allosteric regulation of PRMT3 by exploring diverse compound libraries to identify potential inhibitors that would modify its cellular function. This computational approach involves the generation of pharmacophore hypothesis with 3D-QSAR validation, followed by virtual screening, docking and dynamic simulations to identify potent bioactive compounds as PRMT3 inhibitors. Virtual screening of natural products and PRMT specific libraries revealed the top lead molecules as cladribine, capecitabine, gefitinib, D175-0195, F602-1150 and F1361-0042. However, further validation studies are warranted to confirm these findings.

PRMT3

allosteric inhibition

pharmacophore hypothesis

virtual screening

ADME

Epigenetic regulation refers to the modification of gene expression without changing the DNA sequence. One of the major epigenetic mechanisms is histone methylation, mainly regulating the chromatin restructuring and RNA transcription [1–3]. Protein arginine methyltransferases (PRMTs) are enzymes that methylate arginine residues which serve as key epigenetic markers regulating gene transcription. Specifically, PRMTs synthesize methyl arginine on nucleosomes after being recruited into chromatin remodeling complexes and these PRMTs have been implicated in various cellular processes, including transcription, signaling pathways, splicing, cell cycle, and epithelial-mesenchymal transition (EMT) [4–7]. The methylation status of an arginine residue in histones determines whether the transcription process is activated or suppressed [8].

PRMTs catalyze the transfer of methyl groups from S-adenosylmethionine (SAM) to the guanidino nitrogen atoms of arginine residues in histones and non-histone proteins. PRMTs are divided into three types, based on whether they promote monomethylation (MMA), asymmetric (ADMA), or symmetric dimethylation (SDMA) of arginine [9]. These three derivatives are present on a multitude of diverse protein classes in the cytoplasm, nucleus, and organelles of cells. When two methyl groups are located on one of the terminal nitrogen atoms of the guanidino group, it leads to the formation of the asymmetric dimethylarginine derivative (ADMA). When the methyl group is placed on each of the terminal guanidino nitrogen atoms, it produces the symmetric dimethylarginine derivative (SDMA), and when a single methyl group is found on the terminal nitrogen atom, it is known as the monomethylated Arginine (MMA). In humans, nine PRMTs so far identified are classified into three types as type I arginine methyltransferases consisting of PRMT1, PRMT2, PRMT3, PRMT4, PRMT6, and PRMT8, type II consisting of PRMT5 and PRMT9 that catalyze the formation of ADMA and SDMA respectively. The type III enzyme consists of PRMT7 that only catalyzes the formation of MMA [10, 11].

PRMT3 is a type-I protein arginine methyltransferase that catalyzes the formation of ω-mono- or asymmetric dimethyl arginine. PRMT3 has a unique substrate-binding C₂H₂ Zn finger domain at its N-terminus (Fig. 1) and a catalytic domain at its C-terminus [12]. PRMT3 is widely expressed in human tissues and plays a critical role in many cellular processes. For instance, PRMT3 is crucial for the maturation of ribosomes by regulating ribosomal protein S2(rpS2) methylation, which is critical for the proper maturation of the 80S ribosome. Overexpression of PRMT3 increases H4R3me2a levels in cells, and methylates histone H4 peptides in the nucleus [13]. In addition, PRMT3 is involved in dendritic spine maturation and is crucial for the regulation of various physiological and pathological processes [14]. PRMT3 also mediates the preventive effect of irisin against oxidative stress and lipogenesis. PRMT3 is known to interact with many proteins, such as HMGA1, PABP2, TPI1, MYO18B, and tubulin alpha1c, but the specific molecular mechanisms underlying these interactions remain unclear [15]. Depletion of PRMT3 in mice resulted in small animal size accompanied by reduced methylation of rpS2, while PRMT3 plays a critical role in oculopharyngeal muscular dystrophy by methylating PABPN1 [16, 17]. Specifically, PRMT3 interacts with ALDH1A1 and inhibits its activity by negatively regulating retinoic acid signaling [10]. Moreover, PRMT3 contributes to lipogenesis in hepatoma and non-alcoholic fatty acid liver disease by colocalizing with LXRα in the cell nucleus [18]. PRMT3 levels are enhanced in the myocardial tissues of patients with atherosclerosis [19–20].

The potential role of PRMT3 in cancer is currently being extensively explored. Studies have shown that PRMT3 is overexpressed in various types of cancer such as breast, colorectal, lung and ovarian cancers. PRMT3 mediates metabolic reprogramming and cellular proliferation through methylating R248 of GAPDH, double blockade of GAPDH and mitochondrial respiration could be a novel strategy for the treatment of PRMT3-overexpressing pancreatic cancer [21, 22]. Furthermore, research suggests that targeting PRMT3 could potentially be used as a therapeutic strategy for treating various types of cancer. However, the activity of PRMT3 is not only regulated by its catalytic domain but also by allosteric mechanisms that control its enzymatic activity. Understanding the allosteric inhibition of PRMT3 is a potential therapeutic strategy for the development of new therapies for cancer treatment [23]. Allosteric modulators bind to distal binding sites on PRMT3 and modulate its enzymatic activity. A recent study has identified a potent, selective, and cell-active allosteric inhibitor of PRMT3 [24, 25]. The conformational change of the αC-helix forms a pocket adjacent to, but different from, the ATP-site in which the allosteric inhibitor binds. Further research is needed to explore the full extent of the potential of allosteric inhibition of PRMT3 in cancer treatment. This approach could be particularly promising for colorectal cancer, as PRMT3 is involved in promoting tumorigenesis by methylating and stabilizing HIF1α in colorectal cancer, which is necessary for its stabilization and oncogene function [26, 27]. Hence, targeting PRMT3 in cancer treatment could be a promising therapeutic strategy, and could be combined with other therapies to enhance anti-cancer activity. However, further research is needed to fully understand the mechanisms of action of PRMT3 in cancer cells and to develop effective PRMT3-targeted therapies [28–30]. With this insight, the present study aims to identify potential allosteric inhibitors of PRMT3 using the known allosteric inhibitor SGC707 based pharmocophore generation, 3D-QSAR validated model and virtual screening of diverse compound libraries along with molecular simulations.

2.1 Microarray dataset analysis for DEGs:

The microarray dataset GSE41011 was selected from Gene Expression Omnibus (GEO) (http://www.ncbi.nlm.nih.gov/geo/) which includes the Transcriptome data of Colorectal Cancer patients on Phalanx Human One Array platform. The dataset contains a total of 31 samples with 19 samples from CRC patients of different stages and 12 distant normal samples as depicted in Table 1. From the selected data, raw gene expression was defined using the online statistical tool GEO2R. All raw files of these collected samples have undergone the normalization to remove the ambiguous probes. Consequently, the gene IDs were retrieved and validated between the four stages of colorectal cancer by defining the parameter as log fold-change > 1.0 and p value < 0.05. Analysis was carried out to identify differentially expressed genes (DEGs). Eventually, the upregulated and downregulated genes were extracted and Venn diagram was exploited to explore the common DEGs [31].

Table 1

PRMT3 allosteric site parameters
S.No	Parameters	Score/Volume
1	Score	0.676
2	Pocket Volume	336.244 (Å³)
3	SASA	208.784
4	Pocket Druggability score	0.757

2.2. Structural Analysis of PRMT3:

2.2.1 Protein preparation:

X-ray crystallographic structure of human PRMT3 was obtained from the protein databank (https://www.rcsb.org/) with accession ID 3SMQ. From the X-ray crystallographic data, the protein preparation wizard of the Glide module, Schrödinger Suite, 2018-4 was utilized to resolve specific drawbacks such as missing hydrogen atoms in the protein [32]. The bond ordering in the 3D structure were assigned, any missing hydrogen atoms were allocated and proper ionization states were produced using the default OPLS-3e force field. The grid box was defined using “Receptor Grid Generation” by attributing a point from where a cubic grid box was extended in size. As a net result, the restrained minimization was performed till the heavy atoms converged to an RMSD of 0.3 Å for the execution of the optimal docking [33]. The Glide grid file was created using the receptor grid generating panel within the SAM binding site of PRMT3 (217–531 residues), revealing the x, y, z coordinates as 0.8, -5.26 and 14.86 respectively.

2.2.2 Allosteric site prediction:

Allostery is a biological process that regulates the activity by transmitting the effect of binding at a distant site other than the active site. The identification of the allosteric binding site was attained using the Allosite-Pro server (http://mdl.shsmu.edu.cn/AST/) [34]. Revalidation of the allosteric site was done with PASSer (Protein Allosteric Sites Server) (https://passer.smu.edu), which detects and predicts the probabilities of protein pockets being allosteric sites by normal mode analysis (NMA). For each protein structure, PASSer applies FPocket [35] to split it into several pockets [36]. In this study, the protein PRMT3 (3SMQ) served as the baseline for allosteric site prediction, the algorithm predicts the protein pockets and the highest pocket score was selected for further analysis.

2.3 Preparation of ligand libraries:

Selleckchem provides bioactive compound libraries containing nearly 2576 natural compounds (https://www.selleckchem.com/).These selected compounds were loaded into the Maestro wizard of the Schrödinger suite. The conformers related to minimized structures were filtered through a relative energy window of 10 kcal/mol and a minimum atom deviation of 1.00 Å. Ionization states were generated for each ligand structure and created at a physiological pH of 7.2 ± 0.2. Low-energy 3D conformers with the best bond lengths and angles were generated for each two-dimensional structure. For each ligand structure, ionization states were created at a physiological pH of 7.2 ± 0.2. All other variables were retained at default when the ligands were minimized using an improved potential liquid simulation (OPLS3e) force field [37].

Chemdiv database (https://www.chemdiv.com/) has target-specific PRMTs library containing 27,102 compounds. This reference database includes more than 100 small molecule compounds with reported affinity against PRMTs. The focused library has been collected based on privileged substructure searching including core-head containing analogues, 2D-topological pharmacophore modeling, isosteric morphing and structure similarity searching as well as diversity sub-set. Life Chemicals (https://lifechemicals.com/screening-libraries) Natural Product-like compound library has a proprietary collection of dedicated 14,600 synthetic compounds similar to natural ones, employing similarity search and Substructure Search. These screening compound collections have been recognized to be extremely useful tools for high throughput screening and high content screening programs. These selected compounds were loaded into the LigPrep Maestro wizard of the Schrödinger suite and evaluated for the ligand efficacy [38].

2.4 Generation of Pharmacophore modelling :

A pharmacophore model defines the positioning of molecular and functional group’s characteristics of a ligand in a directive to build a given biological response. Pharmacophore models emerge to find new compounds that should satisfy the necessities of the pharmacophore, which would have a high prospect to be biologically active entity. In this study, SGC707 compound, known inhibitor of PRMT3, was used to design the pharmacophore model. The pharmacophore was built employing the PHASE module 11.8 of Schrödinger Maestro, 2018 [39]. There are six pharmacophore features that are used to define the chemical features of the ligands such as H-bond Acceptor (A), H-bond Donor (D), hydrophobic group (H), negatively charged group (N), positively charged group (P) and Aromatic Ring (R).

2.5 Construction of 3D QSAR model:

Pharmacophore models are the commencement of three-dimensional quantitative structure-activity relationships (3D-QSAR). QSAR modeling was performed using PHASE module 11.8 as well the selected compounds from the chemical entities were employed to develop an atom-based 3D-QSAR model created on developed pharmacophoric maps as a backbone with a partial least-square (PLS) regression. In addition, to evaluate the significance of the biological activities of the model the statistical parameters such as R² (coefficient of determination), SD (standard deviation of regression) and Q² (cross-validated coefficient) were determined [40].

2.6 Virtual screening and Induced Fit docking:

Chemdiv PRMTs library, Selleckchem natural compound library and Lifechemicals natural compound-like library were virtually screened using the best fit pharmacophore model. In this study, DDHRR_1 was the best hypothesis based on which query search was executed to find hits matching the pharmacophoric features of the selected hypothesis by means of the search for match option in PHASE module. The virtual screening was performed using three filtering methods namely, high throughput virtual screening (HTVS), standard precision (SP) and extra precision (EP/XP).

Initially, 10% of the high-scoring compounds were used for the high-throughput virtual screening mode of Glide and 10% of the top-scoring ligands were additionally subjected to Glide SP docking. Likewise, 10% of the top-scoring leads from Glide SP were retained and all the ligands were subjected further to Glide XP docking. Docking and scoring with XP is a more dynamic and distinct method that is executed on the top compounds screened from SP [41]. The 27,102 PRMT inhibitor compounds, 2576 natural compounds and 14,600 natural product-like library were virtually screened using the SP and XP methods based on the above steps. Additionally, the top selected compounds were assessed by performing IFD to precisely find and enhance ligand docking. The approach starts with constrained receptor minimization and accomplishes the first Glide docking of the ligands. The best-receptor structures were identified and the final scoring was analyzed on energy-minimized poses, and the compounds with the best-docked pose and the highest negative score values were considered [42].

2.7 Prime MM-GBSA Energy Calculation :

Prime Molecular Mechanics/Generalized Born and Surface Area (MM-GBSA), the post-docking binding affinity calculation was evaluated using Schrödinger module v11.8. The binding affinities of the complexes were obtained using the OPLS3e force field, a VSGB polar solvation model and a non-polar solvation term with minimized solvation method. The binding free energy and affinity were deliberated using the following formula:

ΔG_(bind) = ΔG_(solv) + ΔE_(MM) + ΔG _(SA)

where E_MM is the energy difference between protein–ligand complex and sum of the energies of protein and ligand. ΔG_solv is the energy difference between the solvation energy of complex structure and the sum of the corresponding energies of protein and ligand. ΔG_SA is the difference between the surface area energies of the receptor–ligand complex and the energies of receptor and ligand [43]. Finally, the retrieved structures with good binding free energies were further examined for ADME/T studies and molecular dynamics simulation runs.

2.8 ADME and Toxicity prediction:

The absorption, distribution, metabolism, excretion, and toxicity (ADME/Tox) profile of small molecules predicts the toxicological and pharmacological features of drug candidates, particularly in the early phases of development. In silico models enhance ADME/Tox pre-dictions and physico-chemical parameters of the retrieved hits. The drug-like behavior of the compounds is normally predicted through the analysis of pharmacokinetic profile of the compounds. Physically significant descriptors and pharmaceutically relevant properties of all the test compounds such as molecular weight, logP, H-bond donors and H-bond acceptors were analyzed in accordance with Lipinski’s rule of five [44]. Finally the selected compounds were filtered through Lipinski’s rule to ensure the properties of pharmacokinetics.

SwissADME online tool (http://www.swissadme.ch/) brings together the most relevant computational methodologies to yield worldwide assessment of small molecule pharmacokinetics profles [45]. The pharmacological properties of the top lead molecules were assessed based on different parameters and the bioavailability radar was generated.

2.9 Molecular dynamics Simulations:

The stability of the selected compounds docked into the allosteric site of PRMT3 was evaluated using the Desmond V5.6 of Schrodinger suite of 2018-4 for 100 ns with the force field OPLS-2005. The structure was generated and solvated using the TIP4P model by orthorhombic water box parameters as well as neutralizing with counter ions from each edge of the box [46]. Before the MD simulations, the complex charge was electrically neutralized as well as the system was minimized with their energies by series of heating and equilibrium processes. The selected complexes were energy minimized by the protocol of steepest descent minimization for the maximum extent followed by a cascade of molecular simulations. The above mentioned steps were taken into account at temperature of 300K with a NVT ensemble by Nose–Hoover thermostats, whereas the pressure was maintained at 1.01325 atm using Berendsen thermostats method [47]. Eventually the salt concentration of the system was set in a specified state as 0.15M Na+/Cl − for the neutralizing components in a dynamics system. The generated MD trajectory files were examined in order to determine the equilibrium of dynamics over the period of time. The stability of the complexes was determined in terms of the root-mean-square deviation (RMSD) and root-mean-square fluctuation (RMSF) for the backbone and side chain of the protein. Moreover, hydrogen bonds also play a vital role in predicting the stable binding in the complex system.

This study is an attempt to identify novel PRMT3 allosteric inhibitors based on the pharmacophore hypothesis generated from the known chemically derived inhibitor SGC707 employing 3D-QSAR model validations and high-throughput virtual screening using Selleckchem, Chemdiv and Lifechemical databases. The screened compounds were further validated as effective drug molecules based on ADMET prediction and the top lead compounds were taken over for further analysis using Molecular dynamics simulations to examine the binding stability of the protein–ligand complexes.

3.1 PRMT3 expression in CRC stages:

Inorder to identify the potent genes associated with the progression of CRC, we selected the dataset GSE41011 and centered on the candidates that were expressed in different stages of CRC. The stage information clinical data (I, II, III and IV) were collected with distant normal non-cancerous samples. First, we identified the differentially expressed genes for each clinical stage of CRC with its distant normal with significant stringent criteria, and subsequently, duplicate gene symbols were removed in each stage. Of these DEGs, 93 were identified in Stage I, 107 in Stage II, 71 in Stage III, and 308 in Stage IV with logFC > 2 and an adjusted P-value < 0.05. Further integrated analysis of DEGs using the Venn diagram (Fig. 2A) revealed that PRMT3 was consistently overlapped and significantly over-expressed in all the four stages of CRC (Fig. 2D). Figure 2B & 2D reveals the stage-wise relative expression of PRMT3. The findings showed that the rates of expression of PRMT3 are correlated significantly with the cancer tumor stage.

The expression of PRMT3 was assessed in a panel of human colorectal cancer samples which includes different clinical stages across various online Omics resources like GEPIA(http://gepia.cancer-pku.cn/), UALCAN (https://ualcan.path.uab.edu/) and cBioportal (https://www.cbioportal.org/) (Fig. 2C,F). The PRMT3 expression was significantly differentially up-regulated across all the tumor stages associated with the cancer cohorts of TCGA, GTex, etc. Our results indicate the heterogeneity of PRMT3 in the CRC progression and might constitute a potent candidate for therapeutic intervention [48].

3.2 Pharmacophore generation

The pharmacophore models were generated using Phase module of Schrödinger suite with the standard set of pharmacophore features employing the known inhibitor SGC707 with IC₅₀ value 31nM as input, along with default parameters. In total, 60 common pharmacophore hypotheses were generated using different combinations of 20 variants. Among the 20 variants, five pharmacophore models DDHHR_1, DDHHR_3, DDHHR_4, DDHHR_5 and DDHHR_6 were selected for further analysis (Supp. Table 1). The scoring procedure provides a ranking of the different hypotheses, allowing us to make rational choices. The scoring algorithm includes the alignment of site points and vectors, volume overlap, number of ligands matched, selectivity, relative conformational energy, and activity [33]. A total of five hypotheses survived the scoring processes were utilized to build an atom-based QSAR model. Hypothesis DDHHR_1 was selected as the best hypothesis based on highest survival score of 5.287 and all other parameters were similar among all models generated (Fig. 3).

3.3 Analysis of 3D-QSAR:

For the accurate prediction of the pharmacophore model, 3D-QSAR model evaluations were executed for the obtained five pharmacophore hypothesis. The statistically significant and predictive ability of the model based on the partial least-square (PLS) regression method was steadily stable at 3 without adverse deviations. DDHHR_1 was discovered to have an acceptable R² value of 0.3414, predictive power Q² value of -0.2921, Pearson R-value of -0.2693, variance (SD) of 1.5044. In addition, the selected DDHHR_1 hypothesis has a P value of 0.0461, pIC₅₀ value of 7.50 and high cross-validated coefficient of correlation indicating its suitability (Supp.Table 2). The atom-based 3D-QSAR model is displayed in Fig. 4.

Table 2

Docking score and binding energy of top hit compounds from various databases with PRMT3
S.No	Compound name/no.	Glide score (kcal/mol)	Glide energy (kcal/mol)	Prime energy	IFD score	Interacting residues
	SGC707	-7.687	-45.364	-11792.50	-597.42	Hbond: R396, E422
Selleckchem Compounds
1	Cladribine	-8.187	-47.709	-11656.55	-591.01	Hbond: D395, E422, T466
2	Capecitabine	-7.032	-45.523	-11648.79	-589.47	Hbond: D389, R396, E422, T466
3	Gefitinib	-5.228	-48.011	-11663.53	-588.82	Hbond: K392, R396 pi-cation: K392, R396, F399
ChemDiv Compounds
4	D585-0607	-5.877	-33.624	-11707.58	-592.25	Hbond: D395, R396, E422 Salt bridge: K392, D395 Pi-cation: K392
5	D175-0195	-5.361	-42.711	-11673.09	-589.12	Hbond: V421, E422, K498 Salt bridge: E422
6	F602-1150	-5.084	-40.018	-11665.32	-589.38	Hbond: K392, D395, R396, E422
Lifechemicals compounds
7	F6127-0048	-5.930	-36.416	-11652.74	-589.84	Hbond: K392, D395, R396
8	F0852-0155	-5.277	-30.485	-11585.95	-586.60	Hbond: K392, R396, E422 Salt bridge: K392
9	F1361-0042	-5.268	-35.408	-11585.09	-586.67	Hbond: D395, E422 Pi-cation: K392

3.4 Virtual screening for the identification of PRMT3 prospective Allosteric Inhibitors from diverse compound libraries:

Virtual screening was performed on the allosteric sites predicted from PASSer and Allosite-Pro employing Chemdiv PRMT library, selleckchem natural product library and lifechemicals natural compound library using DDHRR_1 pharmacophore hypothesis [49]. The allosteric residues in the protein pocket were identified as K392, H393, D395, R396, V387, D389, V420, E422, L424, T466, K498, A467, V501 and L503, while other parameters of the allosteric site (i.e., pocket volume, SASA, perturbation score, and allosite score) are provided in Table 1. To acquire a good cutoff score for the docking study, the compound libraries from Selleckchem, Chemdiv and Lifechemicals were subjected to virtual screening and found to interact with the allosteric site of PRMT3 protein through several molecular interactions. The GlideXP algorithm identified Cladribine as the first hit from the Selleckchem library with a docking score of -8.187 and binding energy of -47.709 kcal/mol. Different types of hydrogen bonding were seen with the D395, E422, T466 residues during the analysis of their interactions. Capecitabine and Gefitinib were the second and third-best ranking compounds from the natural compound library, with glide scores of -7.032 and − 5.228, respectively (Table 2). Both of these compounds also showed strong molecular interactions with the target protein allosteric site. Moreover, Gefitinib displayed strong hydrophobic type interactions ie. Pi-cation reactions were also present in large numbers between the allosteric site’s K392, R396, F399 amino acid residues. Full 3-dimensional and 2-dimensional interactive poses can be seen in Fig. 5.

From the ChemDiv PRMT3 library of compounds, GlideXP ranked D585-0607 as a first hit against the PRMT3 protein with glide score of -5.877. This compound made multiple hydrogen bonding molecular interactions with ASP395, ARG396 and GLU422 of the allosteric site, along with that Vanderwaals type of interactions with the LYS392 and ASP395 were also present between them. Moreover, LYS392 of the allosteric binding pocket exhibited strong hydrophobic interactions of Pi-cation type with the aromatic ring of this compound. The other two top-ranked compounds D175-0195 and F602-1150 had glide docking score of -5.361 and − 5.084, respectively (Supp. Table 3). These compounds are also rich in aromatic moieties resulting in strong hydrophobic interactions between the allosteric site of PRMT3 and the ligands. The second lead compound D175-0195 had strong hydrogen bonding of multiple types with the V421, E422 and K498 of the allosteric site. The third compound F602-1150 also showed strong hydrogen bonded interactions with K392, D395, R396 and E422. The 3-dimensional and 2-dimensional interactive poses of top hit compounds are depicted in Fig. 6.

Table 3

MM-GBSA binding energies (kcal/mol) with ligand-receptor conformational changes accounted in energy terms.
Compounds	MMGBSA -dG-Binding energy	MMGBSA -dG-bind in coulomb	MMGBSA -dG-bind Hbond	MMGBSA -dG-bind Lipo	MMGBSA -dG-bind vdw	MMGBSA -dG-bind (NS)- coulomb
Selleckchem compounds
Cladribine	-251.65	-1.83	-0.53	-0.00	-0.16	-0.00
Capecitabine	-26.34	-3.47	-0.06	-7.66	-18.95	-3.82
Gefitinib	-18.10	-5.75	-0.03	-9.39	-15.28	-2.95
ChemDiv compounds
D585-0607	-30.40	-44.24	-2.73	-10.67	-34.35	-45.42
D175-0195	-41.20	-33.69	-3.45	-14.91	-34.67	-35.70
F602-1150	-37.39	-23.21	-3.16	-12.71	-35.12	-22.05
Lifechemicals compounds
F6127-0048	-29.81	-23.94	-3.15	-7.58	-31.02	-24.94
F0852-0155	-13.78	39.92	-1.72	-6.21	-28.66	38.40
F1361-0042	-33.22	-25.18	-2.72	-12.77	-24.04	-27.02
NS stands for “no strain”

Table 4

ADMET and Druglikeness characteristics of the identified top hit compounds
Compound Name/No.	Physicochemical properties (Lipinski rule of five)					Solubility		Pharmacokinetics
Compound Name/No.	MW (g/mol)	HB donors	HB acceptors	No. of rotatable bond	Consensus log P	Log S (ESOL)	Log S (Ali)	GI absorption	CYP inhibitor
Selleckchem compounds
Cladribine	285.69	3	6	2	-0.26	Very soluble	Soluble	High	No
Gefitinib	446.90	1	7	8	3.92	Moderately soluble	Poorly soluble	High	Yes
Capecitabine	359.35	3	8	8	0.8	Soluble	Soluble	High	No
ChemDiv compounds
D585-0607	335.40	3	4	6	1.48	Soluble	Moderately soluble	Low	Yes
D175-0195	361.78	3	6	8	2	Soluble	Moderately soluble	High	Yes
F602-1150	415.46	2	5	8	2.85	Soluble	Moderately soluble	High	No
Lifechemicals compounds
F6127-0048	288.34	4	3	5	1.25	Soluble	Soluble	High	No
F0852-0155	222.20	4	5	3	-0.12	Very Soluble	Very Soluble	High	No
F1361-0042	290.70	4	6	3	0.19	Very Soluble	Very Soluble	High	No

The Lifechemicals natural product-like compound library is a collection of 14,600 synthetic compounds based on chemical descriptors, natural likeness scoring and similarity to natural scaffolds. The first identified best-ranked compound from this library was F6127-0048 with significantly high molecular contacts with the PRMT3 allosteric site and made conventional hydrogen bonded interactions with K392, D395, R396 and the glide docking score was − 5.930. The other two best-ranked compounds, F0852-0155 and F1361-0042, also showed strong and diverse molecular interactions, with glide scores of -5.277 and − 5.268, respectively (Table 2). The second best-ranked compound exhibited salt-bridge interactions with the K392 of the allosteric site and engaged this residue via conventional H-bonding as well along with R396 and E422. The third hit compound F1361-0042 also had higher molecular interactions and engaged GLU422 twice, along with D395 in hydrogen bonded interaction. This molecule also displayed pi-cation interaction with K392 residue. Figure.7 reveals the three compounds from the lifechemicals natural product-like library in 3-dimensional and 2-dimensional interactive poses, while Table 2 shows their docking score and binding energy.

3.5 Molecular Dynamics Simulation analysis of PRMT3 with selected lead compounds:

Molecular dynamics simulation studies of the three selected Selleckchem complexes revealed that their stability during the 100ns virtual simulation time. The RMSD and the RMSF values of the three complexes can be seen in Fig. 9. The trajectory analysis of these complexes showed that Cladribine and Gefitinib displayed better stability compared to Capecitabine. Cladribine and Gefitinib had various types of contacts during the 100 ns time. The water-bridged assisted hydrogen bond interactions were the most prominent, followed by the conventional hydrogen bonding of the allosteric ASP395, ARG396 and GLU422 residues. LEU503 particularly within the allosteric site of PRMT3 made significant hydrophobic interactions in both the complexes. In addition, Gefitinib complex made hydrophobic contacts with the PHE399 and VAL420 allosteric residues, as shown in the protein-ligand histogram diagram in Fig. 9C. Both these ligands retained and showed the same interactions in the MD studies. The intensity of each of the allosteric residues with Gefitinib shown in Fig. 9C, where K392, D395, R396, F399 and L503 residues were in constant contact with the allosteric site of the PRMT3 along the 100 ns simulation trajectory.

The RMSD fluctuation of Cladribine initiated at 1.2 Å, which is fairly stable, and it remained at a range of 2 Å until the end. The RMSF plot reveals the residue wise fluctuation differences in the protein–ligand complex as well as to identify the conformational changes in the entire simulation period. During the entire simulations, all the protein–ligand complexes exhibited uniform fluctuations in most of the residues except significant displacement at the residues 359–370 close to nearly 6 Å indicating considerable mobility of the residues at those regions and presence of loop between two beta strands. These loop regions often house the orthosteric sites of the enzymes, indicating that allosteric binding considerably changes the conformation of the binding site.

The ligands D585-0607, D175-0195 from the ChemDiv PRMTs library in complex with the PRMT3 allosteric site remained stable during the simulation process. The RMSD of the protein in the complex can be seen in Fig. 10. D175-0195 started to fluctuate at 15 ns time, but the fluctuation was not too high and the change in RMSD was below the 3 Å threshold limit. The protein, RMSD, during the 100 ns simulation, remained like this in the stable form. The ligand, RMSD, also started low, but at about 15 ns time, the RMSD increased and remained like this until 35 ns. It came down after that, and the ligand then stabilized in the allosteric site. The increase of the ligand, RMSD, was due to the torsional movement of some of the movable bonds in this ligand. The interaction analysis of D175-0195 inside the allosteric site revealed that this compound was able to maintain the different types of molecular contacts previously seen in the XP docking studies. The histogram contact diagram can be seen in Fig. 11A, where the major molecular interactions seen in the MD simulation studies were hydrophobic type interactions and water-bridged hydrogen interactions at V421 and E422. The ligand D585-0607 remained stable until the 35 ns mark, after that the RMSD started to fluctuate too much inside the allosteric site (Fig. 10). The RMSD later aligned on the C-alpha backbone of the protein; however, the trajectory analysis suggests that the ligand was in contact with the allosteric site of the PRMT3 during the start of the simulation, but later, some major conformational changes occurred in the ligand structure that resulted in lower interactions. The interaction analysis of the trajectory revealed that this compound engaged in ionic, hydrophobic and H-bond interactions with the allosteric site (Fig. 10A), but its interaction intensity, was low compared to the top identified leads from the other libraries.

Compounds from the Lifechemicals natural-product like library in complex with PRMT3 allosteric site displayed too high RMSD fluctuations than the ligands from other libraries. However, ligand F6127-0048 showed higher intensity of interactions involving ionic, water-assisted hydrogen bonds and hydrophobic bonds at K392, D395, R396, V421 and E422.

3.3 Prime MMGBSA analysis:

The relative binding energies of the selected lead compounds from the various compound libraries with the allosteric site of the PRMT3 were determined using prime MM-GSBA methodology. It provided a variety of protein and allosteric interactions, including Vander Waals, hydrophobic, and coulombic interactions, among others. The MM-GBSA results showed that these allosteric inhibitors have strong binding energies with the allosteric site of PRMT3, and the binding energy terms (i.e., Vander Waals binding energy (dG Bind vdW), lipophilic (dG Bind Lipo), and coulomb energy (dG bind coulomb)) had lower negative values which correspond to stronger binding of the compounds to the allosteric site. Cladridine displayed highly favorable ligand binding with highest binding energy followed by D175-0195, F602-1150 and F1361-0042 (Table 3). Out of the nine compounds screened from three different compound libraries showed diverse scores and the results were tabulated in Table 3. The evaluation of binding free energy under different criterias based on the Prime MM-GBSA methods indicated Cladridine, D175-0195, F602-1150 and F1361-0042 with stronger binding energy and also supported the stability of the molecules with PRMT3.

3.4 ADME/T analysis:

Pharmacokinetics, physicochemical and drug-likeness properties were evaluated through SwissADME online tool, which employs the simplified molecular-input line-entry standard (SMILES) format. The default parameters for orally active drugs must comply with the Lipinski’s rule of five, which asserts that molecular weight (MW) < 500g/mol, hydrogen bond acceptors < 10, hydrogen bond donors and log P values < 5, and the number of rotatable bonds < 10 [29]. The molecular structures of all the nine lead molecules were imported into SwissADME and the different criteria of drug-likeness analysis are depicted in the bioavailability radar in Fig. 8 and table.4. All the nine compounds showed the highest bioavailability score of 0.55 indicating as better oral drug candidates for PRMT3 allosteric inhibition. None of the compounds displayed violationsfor Lipinski’s rule, PAINS and Brent alerts. Compounds Gefitinib, Capecitabine, F6127-0048 and F602-1150 exhibit good drug-like qualities with strong solubility and no excretion problems since it does not interact with pharmacokinetics and does not inhibit CYP enzymes.

Arginine methylation is enzymatically catalyzed by the family of PRMTs that play a central role in epigenetic regulation by synthesizing methyl arginine on nucleosomes, implicated to either activate or repress gene expression depending on the cellular contexts [50–52]. There is emerging interest in understanding the molecular mechanism of PRMT3 associated with disease pathophysiology, also in developing potent PRMT inhibitors. In this extensive analysis, virtual screening for the identification of potent PRMT3 allosteric inhibitors based on pharmacophore and 3D-QSAR analysis has revealed cladribine, capecitabine, gefitinib, F1361-0042 (glucopyranoside derivative) and D175-0195 (oxadiazole-3-carboxamide derivative) as lead molecules. The efficacy of these lead molecules need to be further evaluated through in vitro and in vivo investigations to establish their role as potential allosteric inhibitors.

Wu K, Niu C, Liu H, Fu L (2023) Research progress on PRMTs involved in epigenetic modification and tumour signalling pathway regulation (Review). Int J Oncol 62(5):62. https://doi.org/10.3892/ijo.2023.5510.
Blanc RS, Richard S (2017) Arginine Methylation: The Coming of Age. Mol cell 65(1): 8–24. https://doi.org/10.1016/j.molcel.2016.11.003.
Fuhrmann J, Clancy KW, Thompson PR (2015) Chemical biology of protein arginine modifications in epigenetic regulation. Chem Rev 115(11):5413–5461. https://doi.org/10.1021/acs.chemrev.5b00003.
Bedford MT, Clarke SG (2009) Protein arginine methylation in mammals: who, what, and why. Mol cell 33(1):1–13. https://doi.org/10.1016/j.molcel.2008.12.013.
Hwang JW, Cho Y, Bae GU, Kim SN, Kim YK (2021) Protein arginine methyltransferases: promising targets for cancer therapy. Exp Mol Med 53(5):788–808. https://doi.org/10.1038/s12276-021-00613-y.
Melachrinou M, Logothetis C, Tzelepi V (2021) The protein arginine methyltransferases (PRMTs) PRMT1 and CARM1 as candidate epigenetic drivers in prostate cancer progression. Medicine 100(36):e27094. https://doi.org/10.1097/MD.0000000000027094.
Xu J, Richard S (2021) Cellular pathways influenced by protein arginine methylation:Implications for cancer. Mol cell 81(21):4357–4368. https://doi.org/10.1016/j.molcel.2021.09.011.
Jarrold J, Davies CC (2019) PRMTs and Arginine Methylation: Cancer's Best-Kept Secret? Trends Mol Med 25(11):993–1009. https://doi.org/10.1016/j.molmed.2019.05.007.
Wolf SS (2009) The protein arginine methyltransferase family: an update about function, new perspectives and the physiological role in humans. Cell Mol Life Sci 66(13):2109–2121. https://doi.org/10.1007/s00018-009-0010-x.
Peng C, Wong CC (2017) The story of protein arginine methylation: characterization, regulation, and function. Expert Rev Proteomics 14(2):157–170. https://doi.org/10.1080/14789450.2017.1275573.
Janisiak J, Kopytko P, Tkacz M, Rogińska D, Perużyńska M et al (2021) Protein Arginine Methyltransferase (PRMT) Inhibitors-AMI-1 and SAH Are Effective in Attenuating Rhabdomyosarcoma Growth and Proliferation in Cell Cultures. Int J Mol Sc 22(15): 8023. https://doi.org/10.3390/ijms22158023.
Frankel A, Clarke S (2000) PRMT3 is a distinct member of the protein arginine N-methyltransferase family. Conferral of substrate specificity by a zinc-finger domain. J Biol Chem 275(42): 32974–32982. https://doi.org/10.1074/jbc.M006445200.
Swiercz R, Person MD, Bedford MT (2005) Ribosomal protein S2 is a substrate for mammalian PRMT3 (protein arginine methyltransferase 3). Biochem J 386(Pt 1):85–91. https://doi.org/10.1042/BJ20041466.
Miyata S, Mori Y, Tohyama M (2010) PRMT3 is essential for dendritic spine maturation in rat hippocampal neurons. Brain Res 1352:11–20. https://doi.org/10.1016/j.brainres.2010.07.033.
Park MJ, Kim DI, Choi JH, Heo YR, Park SH (2015) New role of irisin in hepatocytes: The protective effect of hepatic steatosis in vitro. Cell Signal 27(9):1831–1839. https://doi.org/10.1016/j.cellsig.2015.04.010
Min Z, Xiaomeng L, Zheng L, Yangge D, Xuejiao L, Longwei L et al (2019) Asymmetrical methyltransferase PRMT3 regulates human mesenchymal stem cell osteogenesis via miR-3648. Cell Death Dis 10(8):581. https://doi.org/10.1038/s41419-019-1815-7.
Verma M, Khan MIK, Kadumuri RV, Chakrapani B, Awasthi S et al (2021) PRMT3 interacts with ALDH1A1 and regulates gene-expression by inhibiting retinoic acid signaling. Commun Biol 4(1):109. https://doi.org/10.1038/s42003-020-01644-3.
Kim DI, Park MJ, Lim SK, Park JI, Yoon KC et al (2015) PRMT3 regulates hepatic lipogenesis through direct interaction with LXRα. Diabetes 64(1):60–71. https://doi.org/10.2337/db13-1394.
Couto E Silva A, Wu CY, Citadin CT, Clemons GA, Possoit HE et al (2020) Protein Arginine Methyltransferases in Cardiovascular and Neuronal Function. Mol Neurobiol 57(3):1716–1732. https://doi.org/10.1007/s12035-019-01850-z.
Beltran-Alvarez P, Espejo A, Schmauder R, Beltran C, Mrowka R et al (2013) Protein arginine methyl transferases-3 and -5 increase cell surface expression of cardiac sodium channel. FEBS Lett 587(19):3159–3165. https://doi.org/10.1016/j.febslet.2013.07.043.
Yang Y, Bedford MT (2013) Protein arginine methyltransferases and cancer. Nat Rev Cancer 13(1): 37–50. https://doi.org/10.1038/nrc3409.
Hsu SH, Hung WC (2023) Protein arginine methyltransferase 3: A crucial regulator in metabolic reprogramming and gene expression in cancers. Cancer Lett 554:216008. https://doi.org/10.1016/j.canlet.2022.216008.
Jain K, Jin CY, Clarke SG (2017) Epigenetic control via allosteric regulation of mammalian protein arginine methyltransferases. Proc Natl Acad Sci U S A 114(38):10101–10106. https://doi.org/10.1073/pnas.1706978114.
Eram MS, Shen Y, Szewczyk M, Wu H, Senisterra G, Li F et al (2016) A Potent, Selective, and Cell-Active Inhibitor of Human TypeI Protein Arginine Methyltransferases. ACS Chem Biol 11(3):772–781. https://doi.org/10.1021/acschembio.5b00839.
Kaniskan HÜ, Eram MS, Zhao K, Szewczyk MM, Yang X et al (2018) Discovery of Potent and Selective Allosteric Inhibitors of Protein Arginine Methyltransferase3 (PRMT3) J Med Chem 61(3):1204–1217.https://doi.org/10.1021/acs.jmedchem.7b01674.
Hu Y, Su Y, He Y, Liu W, Xiao B (2021) Arginine methyltransferase PRMT3 promote tumorigenesis through regulating c-MYC stabilization in colorectal cancer. Gene 791: 145718. https://doi.org/10.1016/j.gene.2021.145718.
Zhang X, Wang K, Feng X, Wang J, Chu Y, Jia C, He Q, Chen C (2021) PRMT3 promotes tumorigenesis by methylating and stabilizing HIF1α in colorectal cancer. Cell Death Dis 12(11):1066. https://doi.org/10.1038/s41419-021-04352-w.
Wu Q, Schapira M, Arrowsmith CH, Barsyte-Lovejoy D (2021) Protein arginine methylation: from enigmatic functions to therapeutic targeting. Nat Rev Drug Discov 20(7): 509–530. https://doi.org/10.1038/s41573-021-00159-8.
Hartley AV, Lu T (2020) Modulating the modulators: regulation of protein arginine methyltransferases by post-translational modifications. Drug Discov Today 25(9):1735–1743. https://doi.org/10.1016/j.drudis.2020.06.031.
Copeland RA (2018) Protein methyltransferase inhibitors as precision cancer therapeutics: a decade of discovery. Philos Trans R Soc Lond B Biol Sci 373(1748):20170080. https://doi.org/10.1098/rstb.2017.0080.
Murugesan M, Premkumar K (2021) Integrative miRNA-mRNA functional analysis identifies miR-182 as a potential prognostic biomarker in breast cancer. Mol Omics 17(4):533–543. https://doi.org/10.1039/d0mo00160k.
Halgren TA, Murphy RB, Friesner RA et al (2004) Glide: a new approach for rapid, accurate docking and scoring. 2. Enrichmentfactors in database screening. J Med Chem 47:1750–1759. https://doi.org/10.1021/jm030644s.
Mubarak SJ, Gupta S, Vedagiri H (2023) Scaffold Hopping and Screening for Potent Small Molecule Agonists for GRP94:Implications to Alleviate ER Stress-Associated Pathogenesis. Mol Biotechnol 10.1007/ s12033-023-00685-3. https://doi.org/10.1007/s12033-023-00685-3.
Song K, Liu X, Huang W, Lu S, Shen Q, Zhang L, Zhang J (2017) Improved Method for the Identification and Validation of Allosteric Sites. J Chem Inf Model 57(9):2358–2363. https://doi.org/10.1021/acs.jcim.7b00014.
Le Guilloux V, Schmidtke P, Tuffery P (2009) Fpocket: an open source platform for ligand pocket detection. BMC Bioinformatics 10:168. https://doi.org/10.1186/1471-2105-10-168.
Tian H, Xiao S, Jiang X, Tao P (2023) PASSer: fast and accurate prediction of protein allosteric sites. Nucleic Acids Res 51(W1): W427–W431. https://doi.org/10.1093/nar/gkad303.
Harder E, Damm W, Maple J, Wu C, Reboul M, Xiang JY, Wang L et al (2016) OPLS3: A Force field providing broadcoverage of drug-like small molecules and proteins. J Chem Theory Comput 12(1):281–296. https://doi.org/10.1021/acs.jctc.5b00864.
Schrödinger (2020) LigPrep. Schrödinger, LLC, New York.
Dixon SL, Smondyrev AM, Knoll EH et al (2006) PHASE: a new engine for pharmacophore perception, 3D QSAR model development, and 3D database screening: 1. Methodology and preliminary results. J Comput Aided Mol Des 20:647–671. https://doi.org/10.1007/s10822-006-9087-6.
Vanajothi R, Vedagiri H, Al-Ansari MM, Al-Humaid LA, Kumpati P (2022) Pharmacophore based virtual screening, molecular docking and molecular dynamic simulation studies for finding ROS1 kinase inhibitors as potential drug molecules. J Biomol Struct Dyn 40(8):3385–3399. https://doi.org/10.1080/07391102.2020.1847195.
Selvaraj C, Omer A, Singh P, Singh SK (2015) Molecular insights of protein contour recognition with ligand pharmacophoric sites through combinatorial library design and MD simulation in validating HTLV-1 PR inhibitors. Mol Biosyst 11(1):178–189. https://doi.org/10.1039/c4mb00486h.
Friesner RA, Murphy RB, Repasky MP, Frye LL et al (2006) Extra precision glide: docking and scoring incorporating a model of hydrophobic enclosure for protein-ligand complexes. J Med Chem 49(21):6177–6196. https://doi.org/10.1021/jm051256o.
Zhang X, Perez-Sanchez H, Lightstone FC (2017) A comprehensive docking and MM/GBSA rescoring study of ligand recognition upon binding antithrombin. Curr Top Med Chem 17(14):1631–1639. https://doi.org/10.2174/1568026616666161117112604.
Lipinski CA, Lombardo F, Dominy BW, Feeney PJ (2001) Experimental and computational approaches to estimate solubility and permeability in drug discovery and development settings. Adv Drug Deliv Rev 46(1–3):3–26. https://doi.org/10.1016/s0169-409x(00)00129-035.
Daina A, Michielin O, Zoete V (2017) SwissADME: Afree web tool to evaluate pharmacokinetics, drug-likeness and medicinal chemistry friendliness of small molecules. Sci Rep 7:42717. https://doi.org/10.1038/srep4271736.
Jorgensen WL, Maxwell DS, Tirado-Rives J (1996) Development and testing of the OPLS all-atom force field on conformational energetics and properties of organic liquids. J Am Chem Soc 118(45):11225–11236. https://doi.org/10.1021/ja9621760.
Karthika A, Ramachandran B, Chitra J, Prabhu D et al (2021) Molecular dynamics simulation of Toxin-Antitoxin (TA) system in Acinetobacter baumannii to explore the novel mechanism for inhibition of cell wall biosynthesis: Zeta Toxin as an effective therapeutic target. J Cell Biochem 122(12):1832–1847. https://doi.org/10.1002/jcb.30137.
Wang B, Ding D, Zhang X, Zhang J, Zhang H, Liu X, Chang H, Gao Z, Yu Z (2023) The Role of Protein Arginine Methyltransferases in Pathogenesis and Treatment of Digestive System Carcinoma. Mol Cancer Ther 22(8):926–935. https://doi.org/10.1158/1535-7163.
Faisal S, Badshah SL, Kubra B, Sharaf M, Emwas AH, Jaremko M, Abdalla M (2022) Identification and Inhibition of the Druggable Allosteric Site of SARS-CoV-2 NSP10/NSP16 Methyltransferase through Computational Approaches. Molecules 27(16): 5241. https://doi.org/10.3390/molecules27165241.
Morales Y, Cáceres T, May K, Hevel JM (2016) Biochemistry and regulation of the protein arginine methyltransferases (PRMTs). Arch Biochem Biophys 590:138–152. https://doi.org/10.1016/j.abb.2015.11.030.
Fulton MD, Brown T, Zheng YG (2018) Mechanisms and Inhibitors of Histone Arginine Methylation. Chem Rec 18(12):1792–1807. https://doi.org/10.1002/tcr.201800082.
Krause CD, Yang ZH, Kim YS, Lee JH, Cook JR, Pestka S (2007) Protein arginine methyl transferases: evolution and assessment of their pharmacological and therapeutic potential. Pharmacol Ther 113(1): 50–87. https://doi.org/10.1016/j.pharmthera.2006.06.007.

No competing interests reported.

Tables.docx

Download PDF

Version 1

posted

You are reading this latest preprint version

Identification of potent allosteric inhibitors of PRMT3: pharmacophore based 3D-QSAR modeling and molecular simulations approach

Status:

Version 1

Abstract

Figures

Introduction

Materials and methods

2.1 Microarray dataset analysis for DEGs:

2.2. Structural Analysis of PRMT3:

2.2.1 Protein preparation:

2.2.2 Allosteric site prediction:

2.3 Preparation of ligand libraries:

2.4 Generation of Pharmacophore modelling :

2.5 Construction of 3D QSAR model:

2.6 Virtual screening and Induced Fit docking:

2.7 Prime MM-GBSA Energy Calculation :

ΔG_(bind) = ΔG_(solv) + ΔE_(MM) + ΔG _(SA)

2.8 ADME and Toxicity prediction:

2.9 Molecular dynamics Simulations:

Results and Discussion

3.1 PRMT3 expression in CRC stages:

3.2 Pharmacophore generation

3.3 Analysis of 3D-QSAR:

3.4 Virtual screening for the identification of PRMT3 prospective Allosteric Inhibitors from diverse compound libraries:

3.5 Molecular Dynamics Simulation analysis of PRMT3 with selected lead compounds:

3.3 Prime MMGBSA analysis:

3.4 ADME/T analysis:

Conclusion

References

Additional Declarations

Supplementary Files

Status:

Version 1

Identification of potent allosteric inhibitors of PRMT3: pharmacophore based 3D-QSAR modeling and molecular simulations approach

Status:

Version 1

Abstract

Figures

Introduction

Materials and methods

2.1 Microarray dataset analysis for DEGs:

2.2. Structural Analysis of PRMT3:

2.2.1 Protein preparation:

2.2.2 Allosteric site prediction:

2.3 Preparation of ligand libraries:

2.4 Generation of Pharmacophore modelling :

2.5 Construction of 3D QSAR model:

2.6 Virtual screening and Induced Fit docking:

2.7 Prime MM-GBSA Energy Calculation :

ΔG(bind) = ΔG(solv) + ΔE(MM) + ΔG (SA)

2.8 ADME and Toxicity prediction:

2.9 Molecular dynamics Simulations:

Results and Discussion

3.1 PRMT3 expression in CRC stages:

3.2 Pharmacophore generation

3.3 Analysis of 3D-QSAR:

3.4 Virtual screening for the identification of PRMT3 prospective Allosteric Inhibitors from diverse compound libraries:

3.5 Molecular Dynamics Simulation analysis of PRMT3 with selected lead compounds:

3.3 Prime MMGBSA analysis:

3.4 ADME/T analysis:

Conclusion

References

Additional Declarations

Supplementary Files

Status:

Version 1

ΔG_(bind) = ΔG_(solv) + ΔE_(MM) + ΔG _(SA)