RP-HPLC quantification and separation of isomeric bacosides and exploration of their anti-schizophrenic potential by network pharmacology

Abstract

Schizophrenia is a chronic and heterogeneous neuropsychiatric disorder involving positive, negative, and cognitive symptoms. Antipsychotic drugs prescribed currently are effective in controlling positive symptoms but offer limited benefits for negative and cognitive impairments as they focus on single-target or narrow-mechanism treatment strategies without sufficiently addressing the multi-factorial dysfunction underlying schizophrenia. Therefore, multi-target natural compounds like Bacopa monnieri (Brahmi) are increasingly being explored as potential therapeutic candidates for neurological disorders like schizophrenia. However, despite its reported neuroprotective and cognition-supporting properties, the clinical translation of Brahmi remains limited due to the variability in bacoside content across formulations and insufficient understanding of their potential molecular interactions in schizophrenia-associated pathways. In the present study, a reliable HPLC chromatographic method was developed and validated for the simultaneous quantification of major isomeric bacosides, i.e., bacopaside I, bacopaside II, bacoside A3, bacopaside X, and bacopasaponin C, in multiple batches of different formulations. The developed method demonstrated acceptable linearity, accuracy, and sensitivity, enabling consistent bacoside quantification and revealing formulation and batch-dependent variations. To further explore the potential pharmacological relevance and understand the molecular interactions of these bacosides, network pharmacological analysis was performed. Our analysis revealed 97 common targets for the five bacosides, of which 76 were schizophrenia-specific targets. The serotonergic pathway emerged as a key candidate with 5-HT1A, the serotonin receptor, as the top target for bacosides. Molecular docking studies with 5-HT1A confirmed favourable binding poses of bacosides, and bacopasaponin C had the highest docking score. The 50 ns molecular dynamics simulation of the 5HT1A-bacopasaponin C complex demonstrated stable protein–ligand interactions throughout the simulation period. Overall, this study provides an integrated framework combining analytical chemistry with exploratory computational analysis for the anti-schizophrenic potential of Brahmi.

Integrated framework combining analytical chemistry with exploratory computational analysis for the anti-schizophrenic potential of Bacopa monnieri (Brahmi).

1. Introduction

Positive symptoms, negative symptoms, and chronic cognitive impairment are the hallmarks of schizophrenia, a severe and diverse neuropsychiatric illness with significant functional disability.¹ The positive symptoms of schizophrenia are successfully managed by antipsychotic medications,² but the negative and cognitive symptoms constitute a significant unmet therapeutic need.³ The main targets of antipsychotics are the dopaminergic receptors and, to a lesser degree, serotonergic receptors indicating single- or limited-target focus.⁴ Dopaminergic targets remain central in schizophrenia treatment because brain imaging shows that overactive presynaptic dopamine neurons in the striatum lie at the heart of psychosis.⁵ The role of dopaminergic targets in schizophrenia is limited to the positive symptoms, while its effect is weak on the negative and cognitive symptoms.⁶ The complex and multifaceted neurobiology of schizophrenia, which involves the dysregulation of several interrelated pathways, such as glutamatergic transmission, neuroinflammation, oxidative stress, and neurotropic signalling, is not sufficiently addressed by this reductionist approach.^7,8

Multi-target therapeutic approaches that concurrently modulate several biological networks have great value in schizophrenia. In this regard, a promising yet unexplored source of poly-pharmacological agents is medicinal plants and their bioactive components.⁹ Among these, Bacopa monnieri (L.) Wettst., a medicinal plant widely used in traditional ayurvedic medicine, has drawn a lot of interest for its neuroprotective and cognition-enhancing qualities.¹⁰ Preclinical and clinical research has shown that B. monnieri has antioxidant, anti-inflammatory, and neuro-modulatory properties in addition to improving learning, memory, and attention.¹¹ These pharmacological effects of Brahmi coincide with biological mechanisms linked to the pathophysiology of schizophrenia, especially those that underlie negative symptoms and cognitive impairment.

The main bioactive components of B. monnieri contributing to its therapeutic value are bacosides, a type of triterpenoid saponins.¹² Analytical hurdles in quantification and analysis of bacosides, due to their structural complexity, various isomeric and glycosylated forms, pose significant challenges. In addition, diversity in Brahmi formulations, such as extracts, tablets, capsules, and composites and their bacoside concentrations, further complicates accurate reproducible measurement. Existing bacoside quantification methods^13,14 show limitations in accuracy, reproducibility and application to therapeutic formulations (Brahmi tablet, Brahmi ghrita and Saraswatarishta), as they are limited to crude extracts.

To address this gap, a reliable analytical technique for quantitatively determining main bacosides-bacopaside I, bacoside A3, bacopaside II, bacopaside X and bacopasaponin C-in various Bacopa monnieri formulations is reported in this manuscript.

In addition, network pharmacology studies, along with molecular docking and dynamics, were explored to establish the potential molecular interaction of bacosides, with schizophrenia-specific targets. These studies uncovered schizophrenia-relevant biological targets and signalling pathways that are influenced by bacosides. By combining analytical chemistry with systems and structural pharmacology, this study provides a comprehensive analysis of bacosides, of B. monnieri in the context of schizophrenia.

2. Materials and methods

2.1. Chemical standard and formulation

Bacoside reference standards were purchased from Naturewill Biotechnology Co., Ltd, Sichuan province, China, with certified purity (>98%). Commercial Bacopa monnieri formulations were procured from ISO-9001 registered companies (Brahmi powder (Carmel organics, India), Brahmi ghrita (Kerala Ayurveda, India), Brahmi tablet (Himalaya, India), and Saraswatarishta (Dabur, India)). HPLC-grade solvents like acetonitrile (Avantor, India), methanol (Avantor, India), formic acid (SRL chemicals, India) and salts like ammonium acetate (SRL chemicals, India) were procured and Milli-Q water from facility (BITS Pilani, Hyderabad). All formulations and standards were stored under recommended laboratory conditions until analysis.

2.2. Extraction of bacosides and sample preparation

Extraction was carried out using methanol, ethanol, and ethyl acetate, respectively, based on the solubility profile. Methanol exhibited the most effective recovery and reproducibility for extraction, and consequently, it was chosen as the extraction solvent for the extraction of bacosides from commercial formulations. The methodology for extraction from each formulation is given as follows:

(a) Pure Brahmi powder and tablet: Bacopa monnieri powder and tablet underwent maceration with methanol, maintaining a constant sample-to-solvent ratio of 1 : 10 (w/v). The extraction process involved five consecutive cycles to maximize recovery and to prevent the retention residual. Following each cycle, the extract was collected, combined, and concentrated under reduced pressure utilizing a rotary evaporator. The concentrated extract was then dried in a vacuum oven to yield a fine, dry powder.

(b) Commercial therapeutic formulations: Brahmi ghrita is lipid-based, hence an additional component of n-hexane was added to remove the lipid layer. The final ratio was 1 : 10 : 5 (w/v/v) (ghrita : methanol : n-hexane) in the liquid–liquid extraction step. To obtain the bacosides in the methanolic phase, the methanolic layer was separated, concentrated, and dried in the manner indicated above. Saraswatarishta is a fermented liquid formulation; the sample was carefully weighed and lyophilized before extraction to eliminate ethanol and water content. The dried residue was subsequently extracted with methanol by the maceration method and dried in the manner indicated above.

All the dried extracts were then stored at −20 °C, resuspended with methanol and filtered using 0.22 µm syringe filters for chromatographic analysis.

2.3. Development and validation of the analytical method

An in-house new HPLC-based method with a PDA detector was developed for bacoside detection and quantification from the Brahmi formulations using the mobile phases, and the programme is given as follows:

Equipment	HPLC (Shimadzu)
Detector	PDA
Wavelength	205 nm
Column	Supelco ascentis C18, 250 × 4.6 mm, 5 µm, 100 Å
Mobile phase A	10 mM ammonium acetate with glacial acetic acid, pH 4.5
Mobile phase B	Acetonitrile
Column temperature	25 °C
Autosampler temperature	15 °C
Injection volume	20 µl
Flow rate	0.7 mL min−1

Programme	Time (min)	Mobile phase	Percentage (%)
	0.01	B	20
	3.00	B	32
	75.00	B	40
	78.00	B	80
	80.00	B	20
	90.00	B	20

The five bacoside peaks in the composite standard were verified using this method and compared with individual standards for establishing the retention time. Other system suitability parameters like reproducibility, theoretical plate count and resolution were also accounted for.

The developed method was validated for linearity, accuracy, limit of detection, limit of quantification and method specificity for every bacoside.

2.4. Quantification of bacosides

Using the developed and validated method, as a next step, bacoside concentrations were determined in various commercial formulations. Different amounts of dried extracts of formulations were reconstituted in methanol–water (90 : 10, v/v) and diluted with a diluent to fit in the optimal calibration range of 10–200 µg mL⁻¹. Quantification was done in triplicates by sequential injections based on peak area, calibration curve and dilution factor. The calculated concentrations were normalized and reported as milligrams of bacoside per gram of formulation (mg g⁻¹).

2.5. Statistical analysis

All experiments were conducted in triplicate. Data are presented as mean and standard deviation or relative standard deviation with two-way ANOVA, followed by Tukey's multiple comparisons test (p < 0.05) using GraphPad Prism 8.4.2.

2.6. Network pharmacology analysis

The chemical structures of bacosides were obtained from PubChem-National Library of Medicine and compared with the COA of the reference standards of supplier. SMILES of these structures was used for computational analysis.¹⁵ Potential molecular targets of individual bacosides were predicted using CODD-PRED with Chembl 30 as a model data source.¹⁶ MOLBIOTOOLS was used to identify the common target for all 5 bacosides.¹⁷ To check schizophrenia-specific targets from the predicted bacoside list of targets, the curated disease database Open Target Platform and Venny 2.1 (ref. 18) were used. The schizophrenia-specific targets identified were subjected to STRING v12 and Cytoscape 3.10.4 for protein–protein interaction and pathway identification.¹⁹ SR plot analysis for Gene Ontology (GO), functional annotation, and pathway enrichment analyses (KEGG, Kyoto Encyclopaedia of Genes and Genomes) were performed to elucidate the biological processes, molecular functions, and signalling pathways potentially modulated by bacosides. A compound-target-pathway network was constructed and visualized to illustrate the multi-target pharmacological mechanism. Based on KEGG and Cytoscape analysis, the serotonergic pathway was selected and 5-HT1A receptor emerged as top target for molecular docking analysis.

2.7. Molecular docking study

Three-dimensional protein structure of 5-HT1A was obtained from the PDB database (Aripiprazole-bound serotonin 1A receptor-Gi protein complex: RCSB PDB ID 7E2Z). A protein structure was prepared by energy minimization prior to docking by removing non-essential molecules, adding polar hydrogen atoms, and assigning appropriate charges. Molecular docking was performed with Glide XP module of Schrodinger suite in the Schrödinger Maestro (version 2025-1) under the OPLS4 force field. The protein structure was prepared using the protein preparation Wizard by assigning bond orders, adding hydrogen atoms, optimizing protonation states, and performing restrained minimization. Receptor grid generation was centered at X = 101.2, Y = 118.66, and Z = 116.5 with a van der Waals scaling factor of 1.0 and a partial charge cut-off of 0.25. The optimized ligands were subsequently docked in the XP mode to evaluate the binding affinity and ligand–protein interactions.

2.8. Molecular dynamics simulation

Although molecular dynamics provides important insights into ligand-receptor binding interactions, it does not completely represent the dynamic physiological environment of biomolecular systems. Molecular dynamics simulation was performed to evaluate the stability and confirmation of the 5-HT1A receptor bacopasaponin C complex under physiological conditions. The simulation was carried out using the Desmond module of Schrödinger Maestro suite (version 2025-1) under the OPLS4 force field. The docked complex obtained from Glide XP docking was used as the initial structure for the simulation study. The protein–ligand complex was placed in an orthorhombic simulation box and solvated using the SPC water model. Counter ions (Na⁺/Cl⁻) were added to neutralize the system, and 0.15 M NaCl was maintained to mimic physiological conditions. After energy minimization and equilibration, a 50 ns production run was performed under the NPT ensemble at 300 K temperature and 1 atm pressure using the Nose–Hoover thermostat and Martyna–Tobias–Klein barostat. Trajectory analysis was performed to evaluate the RMSD, RMSF, radius of gyration (rGyr), molecular surface area (MolSA), solvent accessible surface area (SASA), polar surface area (PSA), hydrogen bond interactions, and ligand–protein contact profiles throughout the simulation period. The interaction stability of bacopasaponin C with the 5-HT1A receptor was further compared with the reference ligand aripiprazole.^20,21

3. Results

3.1. Analytical method validation and quantification of bacosides

According to the solubility of the standards, the extraction of bacosides was tried using ethanol, methanol and ethyl acetate. Methanol yielded the best results as depicted in Fig. 1a. The same method with methanol was used for the extraction of bacosides from Brahmi formulations, and is depicted in Fig. 1b. Batch-to-batch variation in the triplicates done for five cycles of extraction was minimal, and the carryover impact was negligible. Brahmi powder showed the highest extraction yield compared to Brahmi tablet, Brahmi Ghrita, and Saraswatarishta (Fig. 1b).

Fig. 1. Extraction yield from Bacopa monnieri across solvents and formulations. (a) Total amount of extract obtained using methanol, ethanol, and ethyl acetate, presented as the extract amount (g). (b) Extract recovery (mg g⁻¹ of formulation) from different Bacopa monnieri preparations, including Brahmi powder, Brahmi tablet, Brahmi ghrita and Saraswatarishta. The values are shown as mean ± SD (n = 3).

The HPLC method developed for bacoside quantification showed good resolution (>1.5), constant retention time, minimal baseline drift, and specificity (m/z) under optimized conditions for all the five bacosides in the composite mixture (Fig. 2). Mass validation of each of these bacoside peaks was done using LC-MS (Shimadzu LC-MS 8040) to confirm the specificity of bacosides (SI Data-1).

Fig. 2. HPLC-PDA chromatogram and peak assignment of major bacosides. Representative HPLC chromatogram acquired using PDA detection at 205 nm (bandwidth = 4 nm), showing the separation of five major Bacopa saponins. The peak corresponds to bacopaside I (R_t = 22.495 min), bacoside A3 (R_t = 25.757 min), bacopaside II (R_t = 26.688 min), bacopaside X (R_t = 29.560 min), and bacopasaponin C (R_t = 31.312 min). The compound identity was verified by LC-MS/MS in the negative-ion mode based on m/z values. The inset table summarizes the retention time, resolution, and observed mass for each compound.

The developed method was validated for accuracy, precision, linearity, LOD, and LOQ. Accuracy and precision were evaluated by triplicate injections (n = 3) at the lowest and highest calibration levels, and the % accuracy and % RSD values remained within the acceptable limits. The method showed good linearity across the targeted ranges with correlation coefficient (r²) between 0.9920 and 0.9997. Low LOD and LOQ values confirmed that the method is adequately sensitive for quantitative analysis (Table 1). The resolution factors are indicated in Table 2.

Table 1. Method validation results for bacosides (bacopaside II, bacopaside I, bacoside A3, bacopaside X, and bacopasaponin C). Accuracy is reported as % recovery at low and high levels, and precision as % RSD (n = 3). Linearity is expressed as r² across the stated concentration range, while LOD and LOQ indicate the sensitivity of the method.

Analyte	Accuracy (%)		Precision		r 2	Linearity range (µg mL−1)	LOD (µg mL−1)	LOQ (µg mL−1)
	High	Low	% RSD (high, n = 3)	% RSD (low, n = 3)
Bacopaside II	94.70	130.26	2.795	2.883	0.9920	9.12–182.32	0.92716	2.54809
Bacopaside I	100.45	117.49	0.341	2.366	0.9997	9.29–185.87	1.07962	2.79566
Bacoside A3	96.98	116.11	0.840	6.368	0.9925	9.52–190.48	1.14182	2.20164
Bacopaside X	97.14	93.90	1.387	1.717	0.9984	9.24–184.84	0.96326	1.69885
Bacopasaponin C	100.68	100.65	3.192	7.576	0.9982	9.46–189.21	1.6664	2.20869

Table 2. Resolution values of major bacosides in standard and different Bacopa monnieri formulations under optimized chromatographic conditions. Resolution values greater than 1.5 indicate adequate chromatographic separation between adjacent peaks.

Resolution	Standard	Brahmi powder	Brahmi tablet	Brahmi ghrita	Saraswatarishta
Bacopaside I	—	—	—	—	—
Bacoside A3	4.829	6.140	5.735	5.851	5.857
Bacopaside II	1.567	1.792	1.705	1.736	1.791
Bacopaside X	4.421	4.462	4.282	4.768	4.961
Bacopasaponin C	2.459	2.487	2.461	2.810	2.869

The concentration of bacosides was normalized and expressed as milligrams per gram (mg g⁻¹) of dried extract. The respective chromatograms for each formulation are shown in Fig. 3. Quantitative data showed that different concentrations of bacosides are present in different kinds of formulations, as shown in Fig. 4. Brahmi powder showed the highest bacoside concentration, while Saraswatarishta showed the lowest bacoside concentration (SI Data-2).

Fig. 3. Comparative chromatographic profiling of Bacopa monnieri formulations. HPLC_PDA chromatograms (205 nm, bandwidth 4 nm) obtained from (a) Brahmi powder, (b) Brahmi tablet, (c) Brahmi ghrita, and (d) Saraswatarishta. The chromatograms illustrate the differences in peak intensity and retention behavior for bacopaside I, bacoside A3, bacopaside II, bacopaside X and bacopasaponin C. The embedded tables provide the compound-wise retention time, resolution, and mass confirmation (navigate mode).

Fig. 4. Quantitative distribution of major bacosides in marketed formulations. Concentrations (mg g⁻¹) of bacopaside I, bacoside A3, bacopaside II, bacopaside X, and bacopasaponin C measured in Brahmi powder, Brahmi tablet, Brahmi ghrita, and Saraswatarishta. Bars represent mean ± SEM. Statistical differences in bacoside levels between formulations were evaluated using two-way ANOVA followed by Tukey's multiple comparisons test, with significance expressed as −log(p-value). The significance threshold was set at p < 0.05.

3.2. Network pharmacology

Target prediction analysis of the major bacosides (bacopaside II, bacopaside I, bacoside A3, bacopaside X and bacopasaponin C) identified multiple putative molecular targets for each compound. The predicted targets were 97 (bacopaside II), 101 (bacopaside I), 99 (bacoside A3), 105 (bacopaside X), and 99 (bacopasaponin C), with 97 common targets shared across the compounds. In parallel, schizophrenia-associated targets were compiled, yielding 7139 targets, which were refined to 6559 targets after removing overlaps. Intersection analysis between bacoside targets and schizophrenia targets identified 76 overlapping targets, indicating potential molecular relevance of these bacosides to schizophrenia. The overlapping target set was then used for protein–protein interaction (PPI) network construction and downstream enrichment analysis. The PPI network showed a highly interconnected structure with several hub targets displaying high degree and centrality (Fig. 5).

Fig. 5. Target overlap and protein interaction network of the Bacopa monnieri bacosides in schizophrenia. (a) Venn diagram showing the shared and unique predicted targets among bacopaside I, bacopaside II, bacoside A3, bacopaside X and bacopasaponin C. (b) Intersection of the Bacopa monnieri-associated targets with the schizophrenia-related genes, yielding 76 overlapping targets. (c) Protein–protein interaction network of the common targets visualized in Cytoscape. (d) STRING-based interaction network illustrating the functional connectivity among the overlapping proteins (organism: Homo sapiens).

GO enrichment analysis indicated significant involvement in G protein-coupled receptor signalling, acetylcholine receptor signalling, and phospholipase C-activating GPCR pathways. KEGG pathways analysis further highlighted schizophrenia-relevant pathways, including serotonergic synapse and cholinergic synapse. The compound-target-pathway network confirmed the multi-target and multi-pathway nature of bacoside action (Fig. 6).

Fig. 6. Functional enrichment of Bacopa monnieri-schizophrenia common targets. (a) Gene-pathway interaction network, (b) GO enrichment analysis and (c) pathway enrichment analysis. These three methods revealed the targets of neurological pathways impacted by bacosides, e.g., serotonergic, cholinergic, calcium and estrogen signaling from (a) and (c) and G-protein-coupled receptor signaling, and neurotransmitter responses in (b); especially the neurotransmitter receptor activity like serotonin.

According to the gene ontology and KEGG pathway analysis, the serotonergic pathway emerged at the top pathway relevant to schizophrenia and their targets were selected for Cytoscape to find out the top target for bacosides to show centrality for the pathway specification (Fig. 7). As the target 5-HT1A came as top candidate, it was selected for the next steps of analysis and validation. Schizophrenia is closely linked with disturbances in serotonergic neurotransmission, particularly involving the HTR1A (5-HT1A) and 5-HT2 receptors, which regulate mood, cognition, perception, and dopaminergic signaling.²² Altered serotonin signaling has been associated with hallucinations, cognitive deficits, and negative symptoms commonly observed in schizophrenia.²³ The serotonergic pathway was selected in this study because many atypical antipsychotic drugs exert their effect through the modulation of serotonin receptors. Bacosides are known for their neuroprotective, antioxidant, and neuromodulator properties, making them promising candidates for targeting these altered pathways. Therefore, investigating the interaction of bacosides with schizophrenia-associated serotonergic targets may help to understand their potential therapeutic role in neuropsychiatric disorders.

Fig. 7. Serotonergic synapse and target network analysis. (a) Schematic of the serotonergic synaptic signaling marking major targets for bacosides. (b) Interaction network of the key serotonergic targets, with HTR1A as a central node. (c) Rank order of the principal network nodes based on interaction relevance.

The serotonergic synapse pathway analysis identified several schizophrenia-associated targets, including HTR1A, HTR2B, and SLC6A4, involved in serotonin-mediated neurotransmission. Protein interaction analysis revealed HTR1A as the central hub gene, indicating its major role in serotonergic regulation and synaptic signaling.²⁴ Other interacting targets, such as APP, PTGS2, CYP2D6, and PRKCD, were associated with neuroinflammation, oxidative stress, and neuronal survival pathways.^25–27 These findings suggest that dysregulation of serotonergic signaling contributes significantly to schizophrenia pathology. These computational findings suggest possible interaction between bacosides and serotonergic targets; however, experimental validation is required to determine whether such interactions occur under biological conditions.

3.3. Molecular docking results

Molecular docking was performed to evaluate the binding affinity and interaction pattern of the major bacosides with the selected target protein (PDB ID: 7E2Z), which is associated with serotonergic signalling implicated in schizophrenia. Docking analysis helps predict the stability and molecular interactions between bioactive compounds and target proteins. The docking outcome is shown in Fig. 8, with computed binding energy values ranging from −4.597 to −6.430 kcal mol⁻¹. Bacopaside II scored the least binding energy of −4.597 kcal mol⁻¹ and Bacopa saponin C scored the highest binding energy of −6.430 kcal mol⁻¹ compared to the standard ligand aripiprazole (−9.1 kcal mol⁻¹). The docking scores of the selected bacosides showed acceptable and stable binding affinity toward the target protein associated with serotonergic signalling. Detailed interaction analysis showed conventional hydrogen bond, van der Waals forces, Pi-sigma, alkyl, Pi-alkyl, carbon hydrogen bond, salt bridge and attractive charge. Comparative docking analysis indicated differential binding strengths among bacosides, suggesting structure-dependent target affinity.

Fig. 8. Docking-based interaction analysis of the Bacopa bacosides with the 5-HT1A receptor (PDB: 7E2Z). Docking outcomes for bacopaside I, bacopaside II, bacoside A3, bacopaside X, and bacopasaponin C against the 5-HT1A receptor (PDB ID: 7E2Z). For each ligand, the figure presents the chemical structure, binding pose with the receptor cavity, and a 2D interaction map highlighting the key residue contacts, including hydrogen bonds, hydrophobic interactions, and van der Waals contacts. Binding energies (kcal mol⁻¹) are reported alongside each compound.

To validate the reliability and predictive accuracy of the molecular docking protocol, a self-docking experiment was performed. The co-crystal ligand was extracted from the protein active site and docked back into its original binding pocket using the optimized docking parameters. The superimposition of the docked ligand and the co-crystal ligand demonstrated a close structural alignment with an RMSD value of 1.55 Å for the 5-HT1A partial agonist, confirming the reliability of the docking protocol (Fig. 9).

Fig. 9. Superimposition of the docked co-crystal ligand and the co-crystal ligand with an RMSD of 1.55 Å for the 5HT1A inhibitor for the validation of the study.

3.4. Molecular dynamics results

To evaluate the potential of bacoside with the highest static docking score, bacopasaponin C, molecular dynamics (MD) simulations were performed to account for receptor flexibility and induced-fit structural adaptations. This dynamic approach evaluated the structural integrity and thermodynamic stability of the bacopasaponin C–protein complex over time under realistic physiological conditions, by tracking the root-mean-square deviations (RMSDs) and fluctuations (RMSFs). Furthermore, the timeline analysis mapped the actual persistence and lifespan of crucial intermolecular interactions and water-mediated hydrogen bonds, validating the complex beyond a single static snapshot. The spatial orientation and important residue interactions that contribute to binding stability were demonstrated to be comparable with the original ligand aripiprazole. Evaluation parameters for the 50 ns molecular dynamics simulation included RMSD, RMSF, radius of gyration (rGyr), molecular surface area (MolSA), solvent-accessible surface area (SASA), polar surface area (PSA), hydrogen bond interactions, and ligand-protein contact analysis (Fig. 10). These findings underline the putative anti-schizophrenia significance of bacosides on 5-HT1A and offer molecular-level support for the network pharmacology predictions. The stability and conformational behaviour of the 5HT1A-bacopasaponin C complex was evaluated through 50 ns molecular dynamics simulation. These parameters were used to assess the structural stability, residue flexibility, compactness, and intermolecular interactions throughout the simulation period. The results are provided in SI Data-3.

Fig. 10. (a) RMSD of the 5HT1A-bacopasaponin C complex. (b) RMSF of the 5HT1A-bacopasaponin C complex. (c) Histogram of the ligand-protein contacts of 5HT1A-bacopasaponin C. (d) Ligand protein contacts with amino acids for 5HT1A-bacopasaponin C. (e) Radius of gyration, molecular surface area, H-bond contacts, and SASA of the 5HT1A-bacopasaponin C complex. (f) 2D interactions of the ligand protein contacts of the 5HT1A-bacopasaponin C complex throughout the 50 ns dynamics simulation.

4. Discussion

A major strength of this work is the successful quantification of bacosides in multiple structurally distinct formulations. Most existing analytical studies have been limited to crude plant material or relatively simple solvent extract, which do not represent the formulation complexity encountered in clinical or commercial settings.^28,29 This mismatch has historically contributed to inconsistent efficacy reports and poor reproducibility between batches. Importantly, the present method achieved baseline level chromatographic separation, with resolution values consistently above 1.5, ensuring reliable discrimination of closely eluting bacoside peaks even in interference-rich matrices such as lipid-based ghrita and fermented arista preparations. Another key contribution is the refined approach to bacoside characterization.

The therapeutic translation of Bacopa monnieri has remained challenging despite its long-standing neuro-modulatory reputation. The major limitation has not been the lack of biological relevance, but rather the inability to ensure chemical reproducibility across real-world formulations. This issue becomes even more critical in neuropsychiatric disorders such as schizophrenia, where subtle differences in pharmacological profiles can translate into markedly different outcomes. In this context, the present study offers dual advancement: robust analytical standardization of bacosides across complex matrices and systematic prioritization of schizophrenia-relevant molecular targets through network pharmacology and docking. To further improve the reliability of the computational analysis, docking validation and molecular dynamic simulation studies were incorporated in the present work. These approaches provide additional insights into the stability and consistency of the predicated ligand–receptor integrations under dynamic conditions. Together, these computational approaches provide a hypothesis-generating basis for prioritizing molecular targets and informing the design of future experimental studies; however, they do not constitute evidence of biological activity or therapeutic efficacy, and experimental in vitro and in vivo validation is required before any translational conclusion can be drawn regarding Bacopa monnieri as a neurotherapeutic candidate. Historically, “bacoside A” has often been treated as a single marker compound in the literature, even though it is known to represent a mixture of structurally related saponins.^13,30 This oversimplification has created a translational gap in clinical and pharmacological outcomes that were frequently interpreted without a chemically precise reference point. The present study addresses this limitation by enabling separation and quantification of 5 structurally complex bacosides across multiple batches and formulations. This level of chemical granularity is not merely an analytical achievement, and it is a foundation for pharmacological translation.

However, to achieve pharmacological translation, experimental in vivo and in vitro analyses require the cumbersome study of the action of bacosides on specific schizophrenia targets. Much of the existing literature has emphasized broad biological outcomes such as antioxidant potential, anti-inflammatory effects, or generalized neuroprotection; although these mechanisms are undeniably beneficial in a neurodegenerative or stress-associated context, they remain too nonspecific to explain schizophrenia-related symptom modulation, which is driven by tightly regulated neurotransmitter signaling, synaptic plasticity impairments, and receptor-level dysfunction across multiple interconnected circuits.³¹ To identify schizophrenia-specific targets and to move Bacopa monnieri research away from a single marker mindset towards a more pharmacologically meaningful multi-component standardization framework, a network pharmacology approach was attempted.

The network pharmacology strategy applied in the present study advances beyond descriptive claims by identifying high-confidence targets that are directly embedded in schizophrenia-associated neurobiology, including serotonergic synapse regulation, cholinergic signaling and GPCR-mediated pathways. This transition from “general neuroprotection” to pathway-specific computational prioritization provides a framework for future experimental investigation. Furthermore, it provides a structural rational for the influence of bacosides on specific schizophrenia-relevant phenotypes rather than speculating only on symptomatic improvement. Among the predicted targets, the identification of HTR1A, HTR2A, SLC6A4 and CHRM1 to CHRM4 is particularly notable, as these targets are not peripheral biomarkers but central regulators of neurotransmission and cognition domains profoundly disrupted in schizophrenia.

The serotonergic axis is especially relevant because HTR1A signaling is strongly associated with mood regulation, stress responsive, synaptic modulation, and cognitive processing and serotonergic dysfunction is known to interact with dopaminergic imbalance which remains one of the dominant neurochemical frameworks of schizophrenia.³² The 5-HT1A receptor plays a role in schizophrenia, affecting not only positive features like hallucinations and delusions, but more noticeably the negative and cognitive aspects, such as reduced emotional expression, loss of motivation, and difficulties with planning, focus, and decision-making.³³ Preclinical studies showed that 5-HT1A partial agonists can augment standard antipsychotics and improve overall psychopathology, particularly negative symptoms and some aspects of cognitive impairment.³⁴

Docking of bacosides with 5-HT1A revealed computationally favourable binding poses, suggesting that these compounds may interact with this receptor, while molecular dynamics simulations demonstrated bacopasaponin C and 5-HT1A interactions over 50 ns. Although the docking scores obtained for the bacosides (−4.597 to −6.430 kcal mol⁻¹) were lower than that of the reference ligand aripiprazole (−9.1 kcal mol⁻¹), these findings provide the first computational evidence supporting potential associations between bacosides and the 5-HT1A receptor. These exploratory results are hypothesis generating, and lay the foundation for future in vitro and in vivo studies to establish the biological significance of bacoside and 5-HT1A receptor interactions.

The in silico predictions and exploration of targets by network pharmacology in this study provide a theoretical insight, and can be explored further for definitive mechanistic or translational conclusions. Bacoside separation, quantification and its impact on schizophrenia-specific targets like 5HT1A are the key contributions of this study through analytical chemistry and network pharmacology.

5. Conclusion

The current investigation established a novel, robust and reliable analytical method for bacoside quantification such as bacopasaponin C, bacopaside I, bacopaside II, bacoside A3, and bacopaside X in Bacopa monnieri. The method shows good sensitivity, reproducibility, and accuracy, which is suitable for the routine analysis and standardisation of herbal extracts and formulations containing Bacopa monnieri.

Furthermore, in silico studies involving network pharmacology, molecular docking, and molecular dynamics studies identified multiple schizophrenia-associated targets and pathways that may be relevant to bacoside activity and highlighted 5-HT1A as a computationally prioritised target for future investigation. Overall, these results bridge the traditional medicinal use of Bacopa monnieri with modern computational evidence and provide a hypothesis-generating foundation for future experimental studies investigating the molecular actions of standardised Bacopa monnieri formulations in neuropsychiatric disorders.

Author contributions

Hruta Sundar Swain: investigation, methodology, validation, formal analysis, data curation, writing – original draft and visualization. Balaram Ghosh: supervision, validation, resources and writing – review and editing. Onkar P. Kulkarni: supervision, validation, and writing – review and editing. Vidya Rajesh: conceptualization, methodology, validation, resources, writing – review and editing, visualization, supervision, funding acquisition and project administration. All authors have agreed to the final version of the manuscript.

Conflicts of interest

The authors declare no conflicts of interest.

Data availability

The complete dataset generated and analyzed during this study is available from the corresponding author upon reasonable request.

Supplementary information (SI) is available. See DOI: https://doi.org/10.1039/d6ra04458a.