Introduction

Tardive dyskinesia (TD) was first described by Faurbye et al. in 19641. TD is characterized by involuntary movements, mainly affecting the orofacial region, but can also involve the trunk and limbs. TD includes three clinical forms: classical TD (orofacial TD), peripheral TD (limb-truncal TD), and respiratory TD2. TD typically arises after exposure to dopamine receptor-blocking agents (DRBAs)3,4. In addition to antipsychotic medications, antiemetic drugs such as metoclopramide5, antidepressants6, and calcium channel blockers such as flunarizine7 are also associated with TD. TD may persist or even worsen after DRBAs are discontinued, often becoming irreversible. This condition imposes significant economic burdens8, social isolation4, work impairment9, reduced quality of life10, and higher mortality rates11. The prevalence of TD is lower with second-generation antipsychotics (21%) compared to that with first-generation antipsychotics (30%)3. The pathophysiology of TD remains unclear but may involve oxidative stress12, dopamine receptor hypersensitivity13, and damage to striatal GABAergic and cholinergic neurons13. Managing TD is challenging due to the unpredictable onset, variable presentation, and the necessity of ongoing DRBA treatment for psychiatric or gastrointestinal disorders.

In 2017, the US Food and Drug Administration (FDA) granted breakthrough therapy designations to valbenazine and deutetrabenazine for treating TD, offering new management options for affected patients. The introduction of these VMAT2 inhibitors has rekindled interest in TD treatment. VMAT2 is a protein that transports monoamine neurotransmitters, including dopamine, serotonin, and norepinephrine, into synaptic vesicles in neurons. Valbenazine, a derivative of tetrabenazine, is gradually released after administration and subsequently metabolized to its active form, (+)-α-dihydrotetrabenazine14,15,16. In contrast, deutetrabenazine is a deuterated form of racemic (+/–) tetrabenazine14,17,18. Recent guidelines and expert opinions recommend both valbenazine and deutetrabenazine as first-line therapies for TD18,19. Vitamin E, an essential lipid-soluble antioxidant, protects cell membranes from oxidative damage20. A recent review suggests vitamin E may prevent the deterioration of TD21. Of these treatments, valbenazine and deutetrabenazine are derived from tetrabenazine, which differs from vitamin E in terms of its pharmacology. The common and distinct mechanisms of action for these three agents in the context of TD treatment warrant further investigation.

Network pharmacology is an emerging field that combines systems biology, biological network analysis, and multi-target drug design to predict how drugs can treat specific diseases22. By constructing “drug-pathway-target-disease” networks, this approach uses large datasets to analyze drug interventions, enhancing the understanding of disease mechanisms and therapeutic strategies. Molecular docking, a common virtual screening method, is used to study interactions between receptors and small molecules. Many studies have employed network pharmacology and molecular docking to explore the molecular mechanisms underlying the therapeutic effects of Western medicines on various diseases23,24.

This study utilized integrated database resources to perform a multilevel analysis of valbenazine, deutetrabenazine, and vitamin E. Key methods included target identification, PPI network analysis, GO and KEGG enrichment, and drug-pathway-target-disease network construction, aiming to uncover the shared and distinct molecular mechanisms of these agents in treating TD, as depicted in Fig. 1. By deepening the understanding of TD and its treatment options, this analysis provides a theoretical foundation for developing effective therapies and offers guidance for future advancements in therapeutic strategies.

Fig. 1
figure 1

Workflow of the study.

Full size image

Methods

Target identification for valbenazine, deutetrabenazine, and vitamin E

The chemical structures of valbenazine, deutetrabenazine, and vitamin E were obtained from the DrugBank database (https://go.drugbank.com)25. Targets for three drugs were predicted using the PharmMapper (http://www.lilab-ecust.cn/pharmmapper/)26, Comparative Toxicogenomics Database (CTD) (https://ctdbase.org/)27, GeneCards (https://www.genecards.org/)28, SwissTargetPrediction (http://www.swisstargetprediction.ch/)29, and DrugBank databases. Drug targets were identified after removing duplicates.

Target screening for TD

TD-related targets were identified using the GeneCards, DisGeNET (https://www.disgenet.org/)30, Online Mendelian Inheritance in Man (OMIM) (https://www.omim.org/)31, and Therapeutic Target Database (TTD) (https://db.idrblab.net/ttd/)32. Using “tardive dyskinesia” as a keyword, disease targets were identified after removing duplicates.

Drug-disease target intersection

Venny 2.1.0 (https://bioinfogp.cnb.csic.es/tools/venny) was used to identify intersecting targets between drug-related and TD-related targets. The results were visualized using the bioinformatics platform (http://www.bioinformatics.com.cn/).

PPI network construction and core target screening

Intersecting targets were imported into the STRING database (https://string-db.org) to construct the PPI networks33, restricting the species to Homo sapiens, with a confidence score set to ≥0.4 and excluding disconnected nodes. Cytoscape 3.10.0 (https://www.cytoscape.org/) was used to visualize and analyze the PPI networks34. The CytoHubba plug-in identified the top 10 hub genes using the Maximal Clique Centrality (MCC) method.

GO and KEGG enrichment analyses

Intersecting targets were evaluated by GO and KEGG enrichment analyses using the Database for Annotation, Visualization, and Integrated Discovery (DAVID) (https://david.ncifcrf.gov/)35. GO analysis included biological processes (BPs), cellular components (CCs), and molecular functions (MFs). The top 10 GO terms and KEGG pathways (P < 0.05) were selected for further analysis. Results were visualized using the bioinformatics platform (http://www.bioinformatics.com.cn/).

Drug-pathway-target-disease network construction

Drug-pathway-target-disease networks were constructed and visualized using Cytoscape 3.10.0, highlighting the common and unique core targets and pathways of valbenazine, deutetrabenazine, and vitamin E in treating TD.

Molecular docking verification

Molecular docking was used to predict the binding affinities between three drugs and core target proteins. AutoDockTools 1.5.6 and AutoDock Vina were employed for this process36. Ligand preparation involved downloading structure files for valbenazine, deutetrabenazine, and vitamin E (in sdf format) from PubChem (https://pubchem.ncbi.nlm.nih.gov/) and DrugBank37, followed by conversion to 3D structures using ChemBio3D Ultra 14.0 for energy minimization. These were subsequently processed in AutoDockTools 1.5.6 and saved as pdbqt files. For protein preparation, crystal structures of target proteins were obtained from the RCSB Protein Data Bank (http://www.rcsb.org) and processed in PyMOL 2.4.1 to remove water and ligands38. Hydrogen atoms were added, and charges/atom types were assigned using AutoDockTools, with final pdbqt files saved. Grid preparation involved defining the docking box size and coordinates based on predicted binding sites from DoGSiteScorer (https://proteins.plus/). Molecular docking and binding energy calculations were performed using AutoDock Vina. Finally, Discovery Studio 2021 and PyMOL 2.4.1 were used for interaction analysis and visualization.

Table 1 offers an overview of the tools and databases employed in our research, helping readers better understand the specific methods and supporting materials used.

Table 1 Project overview and resources.
Full size table

Results

Identification of TD-related targets of valbenazine, deutetrabenazine, and vitamin E

Targets for valbenazine, deutetrabenazine, and vitamin E were identified using PharmMapper, CTD, GeneCards, SwissTargetPrediction, and DrugBank databases, yielding 288, 10, 4, 100, and 1 targets for valbenazine; 195, 1, 7, 105, and 1 for deutetrabenazine; and 245, 1941, 917, 22, and 10 for vitamin E, respectively. After removing duplicates, 377 targets for valbenazine, 290 for deutetrabenazine, and 2689 for vitamin E were obtained.

TD-related targets were identified from GeneCards, DisGeNET, OMIM, and TTD databases, with 138, 30, 59, and 1 targets, respectively. After eliminating duplicates, 205 potential TD-related targets were identified. Venn diagrams showed 32 overlapping targets between valbenazine and TD (Fig. 2A), 36 between deutetrabenazine and TD (Fig. 2E), and 62 between vitamin E and TD (Fig. 3A).

Fig. 2: Identification of core targets of valbenazine and deutetrabenazine through protein-protein interaction network analysis.
figure 2

A, E Venn diagrams indicating the common drug-disease targets. A valbenazine and TD; (E) deutetrabenazine and TD. B, F PPI network diagrams for the common targets obtained from the STRING database. B valbenazine; (F) deutetrabenazine. C, G PPI network maps generated using Cytoscape. C valbenazine; G deutetrabenazine. Redder coloration and larger size indicate a higher degree value in the PPI network. D, H Top 10 hub targets selected from the PPI network using the CytoHubba plugin, as ranked using the MCC method. D valbenazine; (H) deutetrabenazine.

Full size image
Fig. 3: Identification of core targets of vitamin E through protein-protein interaction network analysis.
figure 3

A Venn diagrams indicating the common drug-disease targets. B PPI network diagrams for the common targets obtained from the STRING database. C PPI network maps generated using Cytoscape; a redder coloration and larger size indicate a higher degree in the PPI network. D Top 10 hub targets selected from the PPI network using the CytoHubba plugin, as ranked using the MCC method.

Full size image

PPI network construction and core target identification

The 32, 36, and 62 common TD-related targets of valbenazine, deutetrabenazine, and vitamin E were uploaded to the STRING database to construct PPI networks (Figs. 2B, F and 3B). The resulting PPI data were imported into Cytoscape 3.10.0 for further analysis. The PPI network for valbenazine and TD consisted of 32 nodes and 131 edges (Fig. 2C), for deutetrabenazine and TD, 36 nodes and 181 edges (Fig. 2G), and for vitamin E and TD, 54 nodes and 399 edges (Fig. 3C). The top 10 hub targets were identified using the MCC method via the CytoHubba plug-in (Figs. 2D, H and 3D).

GO enrichment analyses

To investigate the molecular mechanisms of valbenazine, deutetrabenazine, and vitamin E in TD, GO enrichment analyses were performed on the common targets. The top 10 BP, CC, and MF terms were displayed in column charts (Fig. 4A–I).

Fig. 4: Results of GO and KEGG pathway enrichment analyses.
figure 4

A–I Bar charts from GO enrichment analyses. J–L Bubble charts presenting the results of KEGG pathway enrichment analyses. A–C, J GO and KEGG pathway enrichment analyses of common targets shared between valbenazine and TD. D–F, K GO and KEGG pathway enrichment analyses of common targets shared between deutetrabenazine and TD. G–I, L GO and KEGG pathway enrichment analyses of common targets shared between vitamin E and TD. A–G The top 10 biological processes. B–H The top 10 cellular components. (C–I) The top 10 molecular functions. J–L The top 10 KEGG signaling pathways. The color and size of each bubble indicate the P-value and gene count, respectively.

Full size image

For valbenazine, 120 BPs, 22 CCs, and 44 MFs were identified. Enriched BP terms included the adenylate cyclase-activating adrenergic receptor signaling pathway, response to xenobiotic stimulus, and xenobiotic catabolic process. Key CC terms were synapse, dopaminergic synapse, and presynaptic membrane, while notable MFs were heterotrimeric G-protein binding, dopamine binding, and monoamine transmembrane transporter activity.

For deutetrabenazine, 120 BPs, 21 CCs, and 47 MFs were identified. Significant BP terms included the G protein-coupled receptor signaling pathway, coupled to cyclic nucleotide second messenger, and response to xenobiotic stimulus. Key CC terms were dendrite, synapse, plasma membrane, and presynaptic membrane, and notable MFs included serotonin binding, G protein-coupled serotonin receptor activity, and dopamine binding.

For vitamin E, 182 BPs, 25 CCs, and 46 MFs were identified. Enriched BP terms included response to xenobiotic stimulus, response to lipopolysaccharide, and xenobiotic catabolic process. Important CC terms included intracellular membrane-bounded organelle, mitochondrion, and extracellular region, while key MFs included heme binding, oxygen binding, and oxidoreductase activity.

KEGG pathway analyses and drug-pathway-target-disease network construction

The analysis identified the top 10 enriched KEGG pathways associated with common targets of valbenazine and TD (Table 2, Fig. 4J). The dopaminergic synapse signaling pathway emerged as the most significant potential pathway, characterized by the lowest P-value and the highest number of enriched core targets. The resulting drug-pathway-target-disease network diagram illustrated the effects of valbenazine on TD, comprising 37 nodes and 77 edges (Fig. 5).

Table 2 KEGG enrichment results for the intersecting targets shared between valbenazine and TD.
Full size table
Fig. 5: The drug-pathway-target-disease network of valbenazine, deutetrabenazine and vitamin E for the treatment of TD.
figure 5

Drug-pathway-target-disease network diagrams relating to the effects of valbenazine (A), deutetrabenazine (C), and vitamin E (E) on TD. The orange inverted triangles represent the drug and disease. The green round rectangles represent the pathways affected by three agents in the treatment of TD, while the red rounded rectangles represent the critical pathways. The blue circles represent the targets of these three agents associated with the treatment of TD. The red triangles represent core targets. B The distribution of the six selected core targets (DRD1, DRD2, DRD3, MAOB, SLC6A3, and SLC18A2) in the dopaminergic synapse signaling pathway. D The distribution of the six selected core targets (DRD1, DRD2, MAOA, MAOB, SLC6A3, and SLC18A2) in the dopaminergic synapse signaling pathway. (F) The distribution of the five selected core targets (AKT1, BDNF, INS, NOS3, and TLR4) in the PI3K-Akt signaling pathway.

Full size image

Similarly, the top 10 enriched KEGG pathways for deutetrabenazine and TD were analyzed (Table 3, Fig. 4K). The dopaminergic synapse signaling pathway again featured prominently and was identified as the most critical pathway with the highest number of enriched core targets. The corresponding network diagram included 44 nodes and 91 edges (Fig. 5).

Table 3 KEGG enrichment results for the intersecting targets shared between deutetrabenazine and TD.
Full size table

Last, the top 10 enriched KEGG pathways associated with vitamin E and TD were examined (Table 4, Fig. 4L). The PI3K-Akt signaling pathway was recognized as the most significant potential pathway due to the largest number of enriched core targets. The corresponding network diagram consisted of 41 nodes and 95 edges (Fig. 5).

Table 4 KEGG enrichment results for the intersecting targets shared between vitamin E and TD.
Full size table

Molecular docking verification

To validate the network pharmacology findings, molecular docking was performed to assess the binding affinity of valbenazine, deutetrabenazine, and vitamin E with selected core targets. Discovery Studio 2021 and PyMOL 2.4.1 were used to visualize the drug-target interactions (Figs. 6 and 7). Based on the degree of hub targets in the PPI network and KEGG enrichment analyses, six core targets were identified for valbenazine and deutetrabenazine, and five for vitamin E. Valbenazine and deutetrabenazine shared five core targets: SLC18A2, SLC6A3, DRD1, DRD2, and MAOB, with DRD3 unique to valbenazine and MAOA unique to deutetrabenazine. The distinct core targets for vitamin E were AKT1, BDNF, INS, NOS3, and TLR4. Details of these potential core targets are listed in Table 5. Docking scores for valbenazine ranged from −7.8 to −8.8 kcal/mol, deutetrabenazine from −6.9 to −8.5 kcal/mol, and vitamin E from −7.1 to −8.7 kcal/mol, indicating stable binding (Table 6).

Fig. 6: Docking patterns of valbenazine and deutetrabenazine interactions with specific core target proteins.
figure 6

A Valbenazine-DRD1; (B) Valbenazine-DRD2; (C) Valbenazine-DRD3; (D) Valbenazine-MAOB; (E) Valbenazine-SLC6A3; (F) Valbenazine-SLC18A2; (G) Deutetrabenazine-DRD1; (H) Deutetrabenazine-DRD2; (I) Deutetrabenazine-MAOA; (J) Deutetrabenazine-MAOB; (K) Deutetrabenazine-SLC6A3; (L) Deutetrabenazine-SLC18A2.

Full size image
Fig. 7: Docking patterns of vitamin E and specific core target proteins.
figure 7

A Vitamin E-AKT1; (B) Vitamin E-BDNF; (C) Vitamin E-INS; (D) Vitamin E-NOS3; (E) Vitamin E-TLR4.

Full size image
Table 5 Information for the 17 core targets.
Full size table
Table 6 Molecular docking scores (kcal/mol).
Full size table

Discussion

TD is a potentially irreversible and disabling adverse effect of DRBAs. Although the FDA has approved VMAT2 inhibitors valbenazine and deutetrabenazine for TD treatment, their precise mechanisms of action beyond VMAT2 interaction (encoded by SLC18A2) remain unclear39,40,41,42. A Cochrane review suggested that vitamin E may help prevent TD progression21, but its exact mechanism is not well understood. This study aimed to explore the molecular targets and pathways of these three agents in TD treatment. Bioinformatics analysis revealed both shared and distinct core targets, suggesting they act through common and unique mechanisms.

Screening for common targets revealed 32 genes shared between valbenazine and TD, 36 between deutetrabenazine and TD, and 62 between vitamin E and TD. These genes may serve as potential targets for TD treatment. From the PPI network and KEGG pathway analyses, we selected six core targets for valbenazine and deutetrabenazine, and five for vitamin E. Valbenazine and deutetrabenazine shared five core targets: DRD1, DRD2, MAOB, SLC6A3, and SLC18A2. Valbenazine uniquely targeted DRD3, while deutetrabenazine specifically targeted MAOA. The five core targets for vitamin E—AKT1, BDNF, INS, NOS3, and TLR4—were distinct from those of the other two agents. These results indicate that valbenazine and deutetrabenazine may exert overlapping pharmacological effects on TD, whereas vitamin E likely operates through different mechanisms.

Abnormalities in dopaminergic neurotransmission, involving VMAT (VMAT1 and VMAT2)40,41,42, dopaminergic receptor genes43, the dopamine transporter (DAT) encoded by SLC6A344, and monoamine oxidase (MAO) activity44,45, may contribute to the pathogenesis and treatment of TD. VMAT2 plays a crucial role in storing and releasing monoamines, such as dopamine, which are implicated in TD42. The VMAT2 inhibitors valbenazine and deutetrabenazine reduce dopamine transport into synaptic vesicles, decreasing dopamine release in the dorsal striatum, a region that regulates movement. This mechanism is thought to mitigate TD, supported by several randomized controlled trials16,17,46,47,48. A review identified associations between DRD1, DRD2, and DRD3 and antipsychotic-induced TD49. Vacuous chewing movements (VCMs), which are unconscious chewing motions without food, often signal neurological or psychological issues and are a symptom of TD. A recent animal study found a positive correlation between VCMs and DAT immunoreactivity in the substantia nigra44. MAO, involved in metabolizing monoamine neurotransmitters, also played a role; chronic VCMs were linked to reduced MAOB activity in rat striatum45. Increased MAO activity can lead to overproduction of hydrogen peroxide50, and dopamine autoxidation can generate dopamine quinones51, which are strong oxidants. These findings, alongside our network pharmacology analysis, support the hypothesis that oxidative stress and dopamine receptor hypersensitivity in nigrostriatal circuits contribute to TD’s pathophysiology and the therapeutic effects of valbenazine and deutetrabenazine.

In addition to dopaminergic dysregulation, immunological mechanisms may play a crucial role in TD development52. AKT1 is downstream of DRD2, which is partially inhibited by antipsychotics. Recently, a genetic study demonstrated the interaction between rs3730358 in AKT1 and rs6275 in DRD2 in patients with TD53. Brain-derived neurotrophic factor (BDNF) possesses neuroprotective and antioxidant properties, mitigating striatal dopamine reductions induced by neurotoxins54. Several studies have reported lower serum BDNF levels in patients with schizophrenia and TD compared to those without TD55,56, suggesting a possible link between neuroimmune dysfunction and TD pathogenesis. Additionally, a study indicated that patients with schizophrenia and TD exhibited reduced movement when treated with low-dose insulin (INS) injections over 3 months57, hinting at potential interactions among metabolic dysregulation, inflammation, and TD severity. Nitric oxide (NO), a physiological free radical, plays a dual role: under normal conditions, it provides neuroprotection, whereas excessive NO production leads to neurotoxicity. Excess NO can form toxic compounds such as reactive nitrogen species, peroxynitrite, and nitrogen dioxide radicals, which induce oxidative stress and inflammatory damage in the central nervous system58. The brain produces NO via two enzymes: neuronal nitric oxide synthase (NOS1) and endothelial NOS (NOS3). Previous research has identified an association between NOS3 gene polymorphisms and TD susceptibility59, further supporting the role of oxidative stress and immune dysfunction in TD pathogenesis. Toll-like receptor 4 (TLR4), a key regulator of innate immunity and neuroinflammatory processes, has been implicated in TD pathophysiology. Evidence suggests that dysregulation of the TLR4 signaling pathway contributes to neuroinflammation and cognitive impairment in schizophrenia patients with TD60. Growing evidence supports that TD results from a complex interplay between oxidative stress, neuroinflammation, and immune system dysregulation, with immunological mechanisms playing a critical role in its pathogenesis. These insights highlight the potential of targeting immune-inflammatory pathways for therapeutic intervention in TD. Notably, Vitamin E, with its potent antioxidant and immunomodulatory properties, may counteract oxidative stress and neuroinflammation, thereby reducing TD progression. A Cochrane review21 indicated that while Vitamin E may not significantly reverse established TD symptoms, it could prevent symptom deterioration by mitigating oxidative damage and modulating immune responses, although the evidence remains of low quality. Additionally, another review20 on Vitamin E supplementation highlighted its role in stabilizing neuronal membranes and scavenging free radicals, further supporting its therapeutic potential in TD. These findings suggest that Vitamin E may provide neuroprotection by reducing oxidative stress-induced neuronal damage and modulating immune responses, making it a promising adjunct therapy for TD.

GO enrichment analyses revealed that, in the context of TD treatment, both valbenazine and deutetrabenazine predominantly target genes involved in dopaminergic synapses, presynaptic and postsynaptic membranes, plasma membranes, and dendrites. Their biological functions include dopamine binding, G-protein binding, and oxidoreductase activity. These GO results align with the proposed etiologies of TD, which involve dopamine receptor hypersensitivity and oxidative stress12,13. Furthermore, KEGG pathway analyses identified the dopaminergic synapse signaling pathway as the most critical common pathway, exhibiting the highest number of core genes, in the effects of valbenazine and deutetrabenazine on TD.

The precise etiology of TD remains unclear. However, one hypothesis is that chronic blockade of dopamine D2 receptors leads to compensatory hypersensitivity of postsynaptic dopamine receptors due to increased receptor density61. This hypothesis does not fully explain why TD persists after the cessation of DRBA use. An alternative hypothesis is that DRBAs block postsynaptic D2 receptors, leading to an excess of unbound synaptic dopamine, which is then reuptaken by the presynaptic neuron via the dopamine transporter. Dopamine autoxidation or enzymatic oxidation via MAO produces reactive oxygen species. Both valbenazine and deutetrabenazine are effective in treating TD by downregulating presynaptic dopamine packaging and release at neuronal synapses62,63. Consequently, the effects of these two drugs on TD may involve targets associated with postsynaptic, synaptic, and presynaptic changes, oxidative stress, and the dopaminergic synapse signaling pathway.

GO enrichment analysis revealed that gene targets of vitamin E in the treatment of TD are predominantly enriched in intracellular membrane-bounded organelles, mitochondria, extracellular regions, and the cytoplasm. Key biological functions include heme binding, aromatase activity, oxygen binding, and oxidoreductase activity. These GO results are in line with previous studies, highlighting the involvement of oxidative stress and inflammation in the effects of vitamin E on TD64,65,66. KEGG pathway analysis identified the PI3K-Akt signaling pathway as the most critical pathway for vitamin E in treating TD. Several studies have demonstrated that activation of the G-protein-coupled estrogen receptor-1 (GPER) enhances the transcription of the ERK1/2 and PI3K/Akt/GSK-3β signaling pathways. This, in turn, can halt the progression of neurodegenerative diseases such as Parkinson’s disease, Huntington’s disease, epilepsy, and TD by reducing oxidative stress, neuroinflammation, and apoptosis67,68,69. Tocotrienols (T3s), a form of vitamin E, particularly γT3/δT3, have demonstrated antioxidant effects and receptor-mediated protection through the GPERβ/PI3K/Akt pathway70.

Blocking the dopamine D2 receptor leads to excessive dopamine metabolism, generating free radicals that contribute to mitochondrial dysfunction, neuroinflammation, apoptosis, and neuronal degeneration. Thus, the effects of vitamin E on TD may be mediated through targets associated with oxidative stress, neuroinflammation, apoptosis, and the PI3K-Akt signaling pathway.

Molecular docking is a widely used tool in computer-aided drug design. We used this approach to further validate the accuracy of our network pharmacology results. The docking scores for the three agents and their corresponding core targets ranged from −6.9 to −8.8 kcal/mol (all <−5 kcal/mol), suggesting strong binding affinities between the ligands and their respective core proteins71. These findings suggest that the effectiveness of the three agents in treating TD may be attributed to their interactions with these core targets. A recent meta-analysis72 suggested that valbenazine and deutetrabenazine are first-line treatments for TD, while vitamin E potentially serves as a second-line treatment. However, given the high cost of valbenazine and deutetrabenazine, these medications remain inaccessible to many TD patients, particularly those from low-income families or developing countries. In contrast, vitamin E is a practical alternative treatment option with its safety profile and affordability. Our study provides computational insights into the molecular mechanisms of these three drugs, further supporting the clinical findings regarding their therapeutic potential. By integrating target identification, pathway enrichment, and network analyses, we demonstrate that vitamin E may exert beneficial effects through mechanisms distinct from or complementary to those of valbenazine and deutetrabenazine. These findings underscore the need to consider both clinical efficacy and accessibility when formulating treatment strategies for TD.

Limitation

This study has several limitations. First, the target data were predicted using various databases, so their reliability depends on the quality of the underlying data. Moreover, the databases used may be more focused on well-established drug targets, potentially overlooking lesser-known or emerging targets that could also play important roles in drug action. Second, network pharmacology mainly relies on qualitative assessments, lacking quantitative analysis of drug dosage effects. Third, how well or poorly vitamin E works compared to valbenazine and deutetrabenazine cannot be ascertained from this study. Our analysis is only a preliminary exploration of the potential mechanisms of these drugs in treating TD. Finally, although this study utilized a data mining approach, further in vivo and in vitro experiments, as well as clinical trials, are required to validate these findings.

Conclusion

This study employed bioinformatics techniques to investigate the pharmacological mechanisms of valbenazine, deutetrabenazine, and vitamin E in treating TD. The findings indicate valbenazine and deutetrabenazine may share common mechanisms by targeting the dopaminergic synapse signaling pathway through core proteins such as DRD1, DRD2, MAOB, SLC6A3, and SLC18A2. However, they also differ in their interactions with specific targets; valbenazine uniquely affects DRD3, while deutetrabenazine targets MAOA. In contrast, vitamin E may operate through a distinct mechanism by regulating the PI3K-Akt signaling pathway via interactions with different core targets, including AKT1, BDNF, INS, NOS3, and TLR4. Our findings provide a foundational understanding for future research and drug development efforts, offering clinicians deeper insights into the pathophysiology of TD. This work may inform targeted therapeutic strategies, ultimately advancing the clinical treatment of TD.