G-protein Coupled Receptors (GPCRs): A Potential Target of Apigenin as a Novel hACE2 Receptor Specific Therapeutic for Impeding Lung Cancer Considering a Group of Missense and Nonsense Mutations in COVID-19 Patients

This research investigates the role of mutation cascades in enhancing COVID-19-related lung cancer fatalities, specifically through analyzing mutations in the ACE2 gene associated with SARS-CoV-2 infections. Notably, a natural flavonoid, apigenin, has been identified as a promising hACE2-specific therapeutic. The study involved detailed examinations of 27 mutations (23 missense and four nonsense) and the molecular interactions between apigenin and hACE2, revealing a binding energy of -8.1 Kcal/mol. Various molecular dynamics parameters suggested stable interactions, while the drug-gene interaction analysis demonstrated that 18 GPCR genes could metabolize apigenin, effectively blocking hACE2 and thereby inhibiting S-protein attachment. The findings propose that apigenin could serve as a targeted therapy for COVID-19-induced lung cancer.

Introduction

Since its emergence in Wuhan in 2019, the COVID-19 pandemic, caused by SARS-CoV-2, has raised global health concerns, leading the World Health Organization (WHO) to declare it a Public Health Emergency of International Concern [1]. The spike protein of SARS-CoV-2 has emerged as a key target for immunotherapy and vaccination [2]. The hACE2 receptor plays a crucial role in the virus’s entry and is interconnected with GPCRs, where therapeutic agents may offer potential treatment options [3]. Historically, medicinal plants have yielded active compounds for medicines, and phytochemicals like hypericin, glycyrrhizin, and apigenin are being explored for their antiviral potential against SARS-CoV-2. Apigenin shows promise as a safe and effective treatment, with anti-inflammatory benefits [4].

Apigenin is one such phytochemical, which has been investigated by multiple studies to have potential medicinal benefits in treating amnesia, depression, insomnia, Alzheimer’s disease, cancer, and diabetes. Apigenin on LPS can stimulate immunity by proliferating human monocytes and macrophages and inhibiting pro-inflammatory cytokines IL-1β, IL-8, and TNF. In addition, apigenin also can inhibit inflammatory chemokines (CCL5) along with adhesion molecules, such as ICAM1 and VCAM1 [5]. Apigenin has been demonstrated to be safe in a study due to having no toxic effects up to 100 mg/kg when administered intraperitoneally to mice and was found to be very effective against SARS-CoV-2 infection [4]. Apigenin has been reported to achieve a significant pharmacological breakthrough in targeting G protein-coupled receptors (GPCRs), as reported previously, considering the drug-gene interaction (DGI) profiles of selective lung cancer genes [6], [7].

This research aimed to explore the genes regulating GPCRs during the binding of the SARS-CoV-2 S protein, focusing on the effects of common mutations in the hACE2 gene on phenotypic changes in lung cancer patients with COVID-19. Besides, the target-specificity of apigenin has been studied as a potential natural flavonoid on the receptor in impeding S protein attachment based on its pharmacokinetic and pharmacodynamic parameters. Finally, a complete DGI network has been established to screen the genes involved in apigenin metabolism for blocking hACE2 before starting the S protein influx mechanism in situ.

Methods

Identification of the Point Mutations in hACE2

Recurring missense and nonsense mutations linked to COVID-19 induced lung cancer within hACE2 gene were discerned through cellular systems, cataloging mutation loci, codon alterations, resultant amino acids, and phenotypic aberrations (Table I). Comprehensive mutation profiling was meticulously performed using the HGMD tool for accuracy [8].

Table I. SARS-CoV-2 Mediated Missense and Nonsense Mutations Involved in GPCRs-Induced ACE2 Receptor Activity-Alteration in Progressing Infection in Human

HGMD codes	Residue positions	Changed codon	Changed amino acid	Phenotypic alterations
CM1822194	30	CGG → CAG	Arg → Gln	↑Immunodeficiency
CM1822193	30	CGG → GGG	Arg → Gly	↑Dysregulation
CM1822195	30	CGG → TGG	Arg → Trp	↑Immunodeficiency
CM1822196	47	CGT → CAT	Arg → His	↑Immunodeficiency
CM155254	49	TGT → TAT	Cys → Tyr	↑B cell lymphocytosis
CM1723772	57	GAA → GAC	Glu → Asp	↑Atopic dermatitis
CM1822197	72	CGA → GGA	Arg → Gly	↑Dysregulation
CM1822198	75	CGG → CAG	Arg → Gln	↑Immunodeficiency
CM1822199	92	TTG → TGG	Leu → Trp	↑Dysregulation
CM1211119	123	GGC → AGC	Gly → Ser	↑B cell lymphocytosis
CM1415045	123	GGC → GAC	Gly → Asp	↑B cell lymphocytosis
CM1211118	134	GAG → GGG	Glu → Gly	↑B cell lymphocytosis
CM1822200	143	AAG → TAG	Lys → Term	Immunodeficiency; dysregulation
CM1511044	150	TGC → TGA	Cys → Term	Severe immunodeficiency
CM1822201	187	CGG → CCG	Arg → Pro	Immunodeficiency; dysregulation
CM1722021	194	CTG → CCG	Leu → Pro	↑Atopic dermatitis
CM1822207	195	GTC → CTC	Val → Leu	Immunodeficiency; dysregulation
CM1822208	362	AAG → GAG	Lys → Glu	Immunodeficiency; dysregulation
CM1722348	394	TCG → TAG	Ser → Term	↑Autism spectrum disorder
CM1822209	408	CGC → CAC	Arg → His	Immunodeficiency; dysregulation
CM1810516	424	CGG → TGG	Arg → Trp	Immunodeficiency; dysregulation
CM1822210	495	CCG → TCG	Pro → Ser	Immunodeficiency; dysregulation
CM1822216	848	CGC → TGC	Arg → Cys	↑Dysregulation
CM1822219	923	TCG → TTG	Ser → Leu	↑Dysregulation
CM133378	945	CAG → TAG	Gln → Term	Severe immunodeficiency
CM1822220	974	CGC → TGC	Arg → Cys	Immunodeficiency; dysregulation
CM1722022	975	CGG → TGG	Arg → Trp	↑Atopic dermatitis

ACE2-Mediated Genetic String Modeling to Screen GPCR Proteins for Viral Infection and Metabolism of Apigenin

The target-mediated gene string was modeled using Cytoscape 3.8.2 to elucidate critical interactions facilitating ‘S-protein’ adhesion and invasion. Drug targets within this network were delineated through GeneMANIA, leveraging coexpression, G-Protein Coupled Receptors Biology (GPCRB) dynamics, exopeptidase, and signaling receptor ligand precursor processing (SRLPP) properties [9].

Molecular Optimization and Docking Analysis

The crystal structure of hACE2 (PDB ID: 1R4L) was obtained from the Protein Data Bank ( https://www.rcsb.org/) and optimized in UCSF Chimera 1.14 [10]. Only the ‘A chain,’ demonstrating the strongest binding with the SARS-CoV-2 S protein [2], was retained, while all extraneous entities (e.g., non-standard residues, water, ligands, ions) were removed.

Apigenin’s pharmacokinetic parameters (PubChem CID: 5280443) assessed via ADMET profiling using SwissADME and Molinspiration Cheminformatics, while pkCSM evaluated its toxicity. Preliminary QSAR data from admetSAR 2.0 was corroborated on the PASS server to explore its anti-infective, antiviral, and antimicrobial capabilities [11]. It’s 3D structure, sourced from PubChem in SDF format, underwent charge optimization using the Gasteiger method [6]. The optimized ligand was converted to ‘mol2’ format for molecular docking studies. Binding affinity between them was predicted via molecular super docking in PyRx 0.8 [12]. The ligand and macromolecule were converted to ‘pdbqt’ format, docking outputs were stored in ‘CSV’ format.

Post-Supramolecular Docking Assessment

Discovery Studio Visualizer v3.0 identified receptor-ligand interactions in the initial phase. PyMOL v2.4.1 refined the analysis and saved it as a ‘pdb’ file. Ligplot+ v2.2.4, leveraging Java scripting, analyzed hydrogen and hydrophobic interactions in the complex [12].

Molecular Dynamics Simulation (100 ns)

CABS-flex 2.0 simulated the ligand-free target macromolecule for 10 ns to analyze its intrinsic dynamics and interactions with ions and water [2]. LARMD simulated the protein-ligand complex for 3.1 ns, assessing some parameters to gauge stability [11]. 100 ns molecular dynamics run in GROMACS tools analyzed parameters including RMSD, RMSF, SASA, MolSA, PSA, B-factor, and Rg, using a 10:10:10 box, Na⁺ ions for neutralization, and a 1.4 Å probe radius [6], [13]. iMODS-NMA assessed deformability, elasticity, eigenvalues, and the B-factor.

Statistical Analysis, and Graphical Representation

After refinement and visualization, molecular dynamics simulation results were statistically evaluated using GraphPad Prism [14]–[16], and R programming (R-studio) [17]–[20].

Results and Discussion

Profiling and Implications of ACE2 Mutations

Twenty-seven missense and nonsense mutations were identified in the target gene, with residues exhibiting heightened mutation frequencies. Codon alterations predominantly induced severe immunodeficiency. Additionally, codon changes triggered B-cell lymphocytosis, a precursor to chronic lymphocytic leukemia (CLL). Four key nonsense mutations were identified, with the TCG → TAG substitution strongly associated with autism spectrum disorder (Table I), echoing prior findings [21].

Identification of GPCR Mechanisms and Gene Impacts in S-Protein Attachment to hACE2 During SARS-CoV-2 Invasion

Gene-string profiling identified ACE2 as intricately linked to twenty-one genes within an operon system, each facilitating the biosynthesis of viral invasion. These mechanisms expedite the influx of foreign particles, amplifying the attachment and internalization of the S-protein during coexpression (Fig. 1A). Notable Genes modulate its activity (Fig. 1B) and some downregulate exopeptidases, impairing cellular defenses and increasing susceptibility (Fig. 1C).

Fig. 1. Molecular string networks representing genes involved in coexpression (A), GPCRB activity (B), exopeptidase (C), and SRLP processing (D) required for S-protein influx mechanism.

Additionally, antibodies targeting the network as COVID-19 biomarkers at the time of global pandemic [22]. The overexpression induced by these mechanisms (Fig. 1B and 1D) to accomplish a successful viral influx mechanism. SARS-CoV-2 mutates transmission pathways to subvert GPCR signaling which means human cells fail to receive any precaution of virulence before the influx takes place [3]. ACE2’s protein-protein interaction (PPI) profile is crucial for viral attachment Interaction source is always very important to describe the PPI and protein-cluster configuration of a functional protein responsible for any disease formation [23]. ROS (Reactive oxygen species)-induced damage in SARS-CoV-2 and lung carcinoma highlights GPCR signaling disruption [24], [25].

Molecular Optimization with Docking and Post-Docking Supramolecular Analysis

The receptor’s energy dropped from −52407.7 to −113291.2 kJ/mol after minimization and verified by YASARA Minimization Server (Fig. 2A).

Fig. 2. llustration of the super-position of the “A chain” of ACE2 receptor based on COACH-D algorithm (A); best predictive docking pose (B); supramolecular docking (C); and the formation of hydrogen bonds and non-covalent hydrophobic interactions between apigenin-ACE2 complex (D).

The predicted binding energy (confidence score of 1.00) for an effective ligand interaction was −6.9 kJ/mol (Fig. 2B). The molecular docking resulted from the predictive binding energy for the flavonoid-protein complex as −8.1 kcal/mol (Fig. 2C). This complex obtained after docking simulation was analyzed and identified hydrogen bonds and hydrophobic interactions. The amino acid residues engaged in hydrogen bonding (green lines) and hydrophobic interactions (red lines) are mentioned (Fig. 2D).

The software was used to visualize the interaction between apigenin and the active residues of IR4L (Fig. 2B and 2C). Additionally, the hydroxyl and carbonyl groups in flavonoid played a prominent role in forming hydrogen bonds, enhancing interaction strength and overall stability [6]. Overall, the ‘apigenin-hACE2’ complex exhibited favorable interaction dynamics and strong binding stability (Fig. 2).

Pharmacodynamic Screening with Molecular Dynamics Simulation (100 ns) and PCA Analysis

Interactions between hACE2 amino acid residues and flavonoid at 3.1 ns are shown in Fig. 3A, with a dendrogram in Fig. 3B illustrating the relationship between structures based on similar residues. Principal Component Analysis (PCA) conducted to evaluate apigenin’s deviation from its initial structure, represented by active and inactive protein frames. The residue correlation matrix for ‘Cα’ atoms over 100 ns in Fig. 3C, highlights correlated (red) and anti-correlated (blue) residues in a 50:50 distribution. The Heatmap in turn gives us a visual representation of the associations and coefficient distances between molecules [26].

Fig. 3. Demonstration of the apigenin-ACE2 receptor complex profile at the 100 ns of MDS (A); a clustering dendrogram derived from principal component analysis (B); a heat map of the molecular dynamics residual cross-correlation matrix (C); PCA cluster analysis with trajectory frames from blue to red in order of time (D), which are recovered by changing conformations from black to red (E); PCA scree plot showing the proportion of variance against its eigenvalue rank, where the first three eigenvectors contribute over 40% of the total variance (D and E).

Fig. 3D shows the trajectory frames, color-coded for active and inactive states and the top three eigenvectors in Fig. 3E reveal that PC1 accounts for 25.06% of the variance, PC2 for 10.45%, and PC3 for 5.76%, collectively explaining 41.3% of the total variation, with remaining components contributing less than 4.3%.The Ramachandran plot was used to validate the stereochemistry of the protein-ligand complex by analyzing torsion angles (phi and psi) [12]. The minimized ligand-bound structure exhibited sufficient accuracy and flexibility, confirming stereochemical stability (Fig. 3A and 3B) was used to distinguish between the structures and to show how their divergence from the parent compound. As mentioned in Fig. 3B, the two clusters diverged at different points. Dihedral phi/psi angle distribution after docking revealed most amino acid residues of 1R4L remained in favorable regions, with 25.06% in PC1, 10.45% in PC2, and 5.76% in PC3 regions (Fig. 3D and 3E). The clustering along PC1 and PC2, accounting for approximately 35% of the structural variance, revealed how structural relationships and conformational changes evolve over time [2].

RMSD; RMSF; Cα; Rg; SASA; MolSA; H-bonds; and PSA

During the simulation, the fluctuation profiles of amino acid residues from 1R4L were examined across 100 ns. RMSD values for the complex ranged from 0.283 Å to 0.147 Å (Fig. 4A and 4B), while the receptor, however, and exhibited significant RMSF fluctuations between 0.054 Å and 0.458 Å (Fig. 4C). Cα values fluctuated from 0.396 Å to 3.517 Å (Fig. 4D), while Rg varied between 2.352 nm and 2.42 nm (Fig. 4E). SASA values varied between 223.893 Ų and 263.812 Ų (Fig. 4F), with a default Water Probe Radius (WPR) of 1.4 Å and MolSA of the complex fluctuated between 236.665 Ų and 244.14 Ų (Fig. 4G). Out of 1001 frames, 946 formed a hydrogen bond, while 55 did not (Fig. 4H) and PSA remained between 188.675 Ų and 200.676 Ų (Fig. 4I).

Fig. 4. Systematic profiling of the receptor RMSD (A); ligand RMSD (B); RMSF (C); Cα (D); Rg (E); SASA (F); MolSA (G); H-bonds (H); and PSA (I) values resulted from the MDS (100 ns).

Lower RMSD values correlate with increased stability, and the obtained RMSD for the apigenin-1R4L complex suggests excellent stability [7] (Fig. 4A and 4B). RMSF fluctuations, representing the average Cα fluctuations, were measured against a reference Cα state over a 100 ns simulation to evaluate apigenin’s effect on 1R4L (Fig. 4C and 4D). In Fig. 4C, the values showed similar peaks and overlapping fluctuations, indicating favorable stability (Fig. 4D). A negative correlation between stability and RMSF was observed, with tighter structural bonding (such as α-helices and beta-sheets) showing higher values, and more stable structures exhibiting lower fluctuations [11]. Rg, SASA was evaluated to determine the compactness and stability of the protein-ligand complex and indicated instability, solvent exposure and flexibility, while lower values suggest a stable, dense structure [6]. The complex exhibited minimal fluctuations, suggesting a stable and tightly packed structure (Fig. 4E and 4F).

A higher surface area during MD simulation indicates ligand dissociation from the binding pocket, leading to reduced complex stability. In contrast, lower atomic-level MolSA values are indicative of a more stable interaction [11]. In this study, the apigenin-hACE2 complex was found to be somewhat stable within a MolSA value range of 236.665 to 244.14 Å2 (Fig. 4G). Hydrogen bonds can be implied that the complex formed 446-539 intramolecular hydrogen bonds over 100 ns, indicating robust stability (Fig. 4H). In recent times, to avoid drug resistance to antiviral drugs, a range of bacteriocins (anti-microbial proteins) have been emphasized to combat SARS-CoV-2 isolated from different probiotic microorganisms [27], supporting immune response processes like opsonization [28], [29].

In this in silico study, flavonoids adhered to Lipinski’s rule of 5’ and exhibited high intestinal absorption (>90%). Despite a negative blood-brain barrier range, the logBB value (0.3 < logBB < −1) indicated minimal permeability [11]. Compounds with PSA < 60 Ų are BBB permeable, while those > 120 Ų exhibit poor permeability. Here, apigenin, with a PSA > 188 Ų (Fig. 4I), demonstrates poor penetration [4].

Deformability; B-Factor; Eigenvalue; and Elasticity Index

The deformability index indicated significant structural deformation when complexed with apigenin at the atomic index n = 596 (Fig. 5A). The Debye-Waller factor (B-factor) values exhibited the highest flexibility, with amplified NMA wave profiles relative to PDB values (Fig. 5B). Conversely, the stiffness of ACE2, particularly in the “A chain” region, ranged from 1 to 5.1, with an Eigenvalue of 3.144597e-04 which directly regulated with the promoted energy demand for protein in the precise regions (Fig. 5C), indicating higher deformability energy. The elasticity index revealed optimal residue connectivity. These findings emphasize the interplay of flexibility, energy, and structural changes during protein-ligand interactions [30] (Fig. 5D).

Fig. 5. Structural profiling of the deformability (A); B-factor (B); Eigenvalues (C); and molecular elasticity (D) based on the electrostatic reinforcements of MDS.

DGI Interactions and Metabolism of Apigenin

The study identified eighteen genes involved in apigenin metabolism, inhibiting protein receptor complexation with the S protein. Inside the remarked area of genes, apigenin primarily serves as the interaction source for these genes, while notably receiving metabolic signals from nine genes. In the GPCRs operon system, forty-seven genes were identified, with eighteen playing a key role in flavonoid metabolism, preventing foreign particles from binding to the protein receptor (Fig. 6).

Fig. 6. Proteins other than the ACE2 receptor are targeted by apigenin simultaneously.

Apigenin effectively blocks the target receptor, based on its pharmacokinetic and pharmacodynamic profiles and a similar type of genetic metabolism is found for the other natural flavonoids such as Myricetin, quercetin, and isovitexin [13]. Even in HER2-positive breast cancer and hepatic fibroblast treatment [6], DGI characterization of apigenin-type flavonoids has been used in both in vitro and in vivo steps.

Recent advancements highlight the potential of oleaginous microbes, including yeast, fungi, and bacteria, as sustainable sources of anticancer lipids and bioactive molecules. These microbial-derived compounds, such as polyunsaturated fatty acids (PUFAs) and glycolipids, exhibit anticancer properties by modulating signaling pathways and inducing apoptosis in cancer cells, akin to the mechanisms of apigenin and other flavonoids [31]. Furthermore, the lipid profiles of these microbes can be tailored for functional food applications, offering a viable alternative to plant-based phytochemicals [32]. Interestingly at present, a group of DFT (density functional theory) approaches such as SCF (self-consistent field), RDG (reduced density gradiant), and eigenvalues have been introduced alongside the classical molecular dynamic simulation in explaining the mechanics working behing the H-bonding and hydrophobic interactions during ligand-protein complexing paving the way of both next-generation molecular drug designing and functional food formulation against cancer [33–35].

Conclusions

This study examines twenty-seven mutations in the ACE2 gene, including four nonsense mutations, which impact various phenotypic expressions like immunodeficiency, B cell lymphocytosis, and autism spectrum disorders in lung cancer patients. Molecular network analysis of GPCRs, including the receptor gene, identified key genes involved in S-protein influx mechanisms. Apigenin, a potential flavonoid therapy, is strongly bound to the hACE2 receptor (−8.1 Kcal/mol), showing promising stability in molecular dynamics simulations. DGI studies also revealed the synergistic impact of specific amino acid residues on apigenin binding.