Alzheimer’s disease (AD) and Parkinson’s disease (PD) are the most rapidly increasing neurological disorders globally, posing significant management challenges owing to the progression of motor and cognitive disabilities in affected individuals [1]. Both neurodegenerative disorders are characterized by intricate molecular mechanisms that impede the development of effective treatments. Aβ plaques, neurofibrillary tangles, gliosis, and neuronal loss are the primary pathological features of AD. Eventually, individuals with AD experience severe memory loss, hallucinations, disorientation, and loss of autonomy [2]. In contrast, α-synuclein aggregation, ferroptosis, and mitochondrial dysfunction are hallmarks of PD. Bradykinesia, rest tremors, rigidity, and loss of postural reflexes are the motor symptoms caused by this molecular process. Furthermore, non-motor symptoms, such as hypotension, constipation, urinary disturbances, sleep disorders, and a spectrum of neuropsychiatric symptoms, further complicate the treatment of PD [3, 4]. Although synthetic agents have shown promise as multifunctional drugs for AD and PD, key limitations, such as pharmacokinetics and safety concerns, persist. Unfortunately, the medications currently available for AD and PD only alleviate symptoms and do not halt neurodegeneration, underscoring the significance of novel drug discovery. Natural products may offer efficient and safe pharmacodynamic properties for the treatment of these challenging neurodegenerative disorders. Consequently, plants and their bioactive compounds have been extensively investigated for their therapeutic potential in various neurodegenerative diseases, including AD and PD [5, 6].

In the last few years, capsaicin (CAP), a prominent constituent of chili peppers and member of the vanilloid family, has gained significant attention within the scientific community due to its wide array of pharmacological effects and diverse biological properties, including pain management, cardiovascular health, anti-inflammatory, and metabolic regulation [7]. Daily CAP consumption worldwide is estimated to be 1.5 mg per person in the United States and Europe, 25 mg per person in India, and 200 mg per person in Mexico [8]. In addition, CAP has been shown in animal studies to have oral bioavailability of 50–90% and to cross the blood-brain barrier (BBB) [8]. This phytochemical compound activates the transient receptor potential vanilloid 1 (TRPV1) receptor, which is implicated in nociception, offers neuroprotection against aggregation of misfolded proteins, and improves mitochondrial function [9]. However, CAP toxicity within the central nervous system may result in seizures, increased activity, confusion, and compromised motor function [10]. Thus, the efficacy of plant-derived chemical compounds is often complicated by their interactions with target molecules, both synergistic and antagonistic [11]. Computational biology and bioinformatics tools, such as pharmacological networks, molecular dynamics simulations, and molecular docking, have emerged as effective methodologies for elucidating these interactions and assessing drug likeness and pharmacokinetic properties [12–14]. Recent advancements in computational biology have further enabled the refinement of in silico study processes through the application of machine learning and deep learning, thereby enhancing the quality of biomedical data interpretation [15, 16]. Although the potential neuroprotective effects of CAP have been previously investigated in vitro and in vivo, this study introduces an innovative approach by employing network pharmacology to identify shared mechanisms between AD and PD. Furthermore, it validates the stability of CAP interactions with its primary targets [sirtuin 1 (SIRT1), estrogen receptor 1 (ESR1), heme oxygenase 1 (HMOX1), and nerve growth factor (NGF)] through 100 ns molecular dynamics simulations and Molecular Mechanics Poisson-Boltzmann Surface Area (MM/PBSA) calculations. Therefore, this network pharmacology study aimed to explore the molecular mechanism of CAP and its targets using bioinformatic tools, to suggest CAP as a possible alternative treatment for AD and PD.

CAP is the primary pungent compound in chili peppers and has been used extensively in traditional medicinal practices for centuries. On the other hand, AD and PD are neuropathological conditions primarily characterized by the accumulation of misfolded proteins: alpha-synuclein in PD and tau protein, along with beta-amyloid peptides, in AD. The presence of these misfolded proteins initiates other pathophysiological processes common to both disorders, including mitochondrial dysfunction, reactive oxygen species (ROS) production, and neuroinflammation. These processes contribute to neuronal loss and synaptic disconnection in the hippocampus in AD, and the degeneration of dopaminergic neurons in the substantia nigra in PD [2, 3]. Thus, investigating the mechanisms of CAP in AD and PD requires a thorough understanding of its molecular interactions and biological mechanisms. This understanding can be facilitated through advanced in silico methodologies.

In this context, although the neuroprotective potential of CAP has been previously investigated [39, 40], this study introduces an innovative approach by incorporating molecular mechanisms and network pharmacology analysis. This approach identifies 25 common targets and 9 hub proteins, with rigorous validation via molecular dynamics simulations for 100 ns and binding free energy (MM/PBSA) calculations. This dynamic validation is essential, as it confirms the structural feasibility and stability of CAP interactions with key therapeutic protein targets such as SIRT1 and ESR1, thereby providing robust structural evidence that was previously unavailable in the literature.

Pharmacological networks represent a multidisciplinary research domain encompassing genomics, proteomics, and biological systems. These methodologies facilitate the integration of data regarding interactions among biological molecules, signaling pathways, and disease networks, thereby elucidating the mechanisms by which natural compounds exert their effects [41, 42]. In this study, we delineated and analyzed the interactions between CAP and its 25 targets within the contexts of the PD and AD pathways. Our GO analysis revealed that these targets are predominantly associated with BP, including rhythmic processes. Previous research has shown that CAP can affect and regulate circadian rhythms and brain activity [43, 44], primarily through the TRPV1 channel; however, TRPV1 was not identified among the 25 genes included in the GO analyses (see Table S1, Table S2, Table S3, and Table S4). Rohm et al. [45] found that CAP regulates the expression of genes such as DRD2, independently of the TRPV1 signaling pathway. In a recent bioinformatic study, DRD2 has been identified as a pivotal gene linking PD to the circadian rhythm [46]. GSK3B is another target implicated in these processes, particularly in neural activities. Moreover, Xu et al. [47] observed that CAP appears to prevent tau protein hyperphosphorylation by enhancing PI3K/AKT activity and inhibiting GSK3B in the hippocampus of a type 2 diabetes rat model, contingent upon the TRPV1 signaling cascade. Thus, it is essential to note that CAP’s impact might be dependent or independent of the TRPV1 signaling pathway. The αvβ3 integrin-IGF1-IGF1R complex and growth factor complex were identified as the principal cellular components affected by CAP in our analysis. There is a paucity of information regarding CAP’s influence on integrins within the IGF1-IGF1R complex; thus far, it is only postulated that CAP can elevate IGF-I levels [48]. Regarding the development factor complex, CAP may promote VEGFA expression and be involved in angiogenesis [49]. In the context of neurodegenerative diseases like AD and PD, this target holds significant importance. VEGFA is directly linked to neuronal excitability, neurotransmission, and synaptic plasticity. Nonetheless, elevated levels of VEGFA can compromise the integrity of the BBB and increase microvascular permeability [50]. These findings underscore the importance of accounting for variables such as dosage, physiological state, and formulation when administering CAP.

Cholinesterase activity and dopamine neurotransmitter receptor activity, coupled with the protein G_i/G_o, are MFs implicated in the CAP mechanism. Acetylcholinesterase (AChE) and butyrylcholinesterase (BChE) are members of the cholinesterase enzyme family, with BChE’s role being less well-defined compared to AChE. While AChE is primarily recognized for degrading acetylcholine in neural synapses, it has also been reported to participate in bone growth and remodeling [51, 52]. Mansalai et al. [53] found that CAP and hydrocapsaicin exert a mixed inhibitory effect on AChE, as indicated by both in vitro and in silico experiments. This has led to the consideration of CAP and its derivatives (capsaicinoids) as possible therapeutic agents for AD [53–55]. Conversely, CAP does not directly interact with dopamine receptors, such as DRD2 and DRD3; however, it can influence dopamine release via TRPV1 [45].

In this research, Reactome pathway enrichment analysis identified that AD and PD genes are associated with dopamine receptors and activation of caspases through the apoptosome. This result matches the previously obtained outcome for MFs. In cancer cells, it is well known that CAP can activate caspases, such as CASP3 and CASP7, which trigger programmed cell death [56, 57]. In this neurological framework, CASP3 has been implicated in the conversion of amyloid precursor protein into amyloidogenic fragments and the early-stage accumulation of caspase-cleaved amyloid precursor protein in AD [58]. In addition, both caspases are integral to the degeneration of dopaminergic neurons, a hallmark of PD [59]. Some studies demonstrated the capacity of CAP to decrease CASP3 expression [60, 61], although there are no studies regarding CASP7. On the other hand, although CAP can inhibit cell death in neurons, some studies suggest that it can also activate the caspase cascade to promote apoptotic processes [62].

Analysis of the protein-protein interaction network identified nine potential proteins (IL6, GSK3B, CASP3, BCL2, ESR1, SIRT1, NGF, IGF1, and HMOX1) for molecular docking with CAP. Molecular docking, also referred to as computationally simulated ligand binding, is a robust technique for predicting interactions between a ligand (small molecule) and a protein at the atomic level. This method holds significant potential in nutraceutical research and has evolved into a powerful tool for drug development. Prior in vitro studies have employed molecular docking analyses in nutraceutical research to provide critical insights [63, 64]. All interactions of CAP with the nine proteins exhibited low Gibbs free energy (∆G) values; however, SIRT1, ESR1, HMOX1, and NGF showed the highest affinity binding. SIRT1 is a member of the sirtuin family of proteins, which are found in subcellular compartments. These proteins are involved in various processes within the nervous system, including axon development, neuronal differentiation, and neurite outgrowth [65]. Moreover, in AD and PD, SIRT1 overexpression may exert a neuroprotective effect by regulating neuroinflammation and autophagy [66, 67]. In a murine model, CAP administration at 2 μg/h for 28 days can induce SIRT1 expression, thereby reducing ROS and decreasing inflammatory cytokines in the paraventricular nucleus [68]. Conversely, CAP can enhance SIRT1 expression by activating TRPV1, thereby facilitating Ca²⁺ influx into human umbilical vein endothelial cells [69]. It has been reported that a moderate and well-regulated increase in calcium influx can serve as a neuroprotective mechanism. Still, it could result in a neurotoxic condition, in AD and PD, when the influx is prolonged and excessive [70]. Additionally, the induction of SIRT1 expression via CAP might promote axon development in embryonic hippocampal neurons [71]. The ESR1 gene encodes estrogen receptor 1, which binds estrogen and helps regulate mood and cognitive function [72]. This protein has been implicated in the neuroprotective process against PD and AD by mediating estrogen to support the lifespan of dopamine neurons and reduce neuroinflammation [73, 74]. Recently, an in vitro study by Pietrowicz and Root-Bernstein [75] showed that CAP directly interacts with ESR1, increasing its binding to estradiol via a TRPV1-independent mechanism at CAP concentrations of 0.001–0.1 mg/mL. Heme oxygenase 1 is an enzyme associated with the expression of the HMOX1 gene, which catalyzes the conversion of heme to carbon monoxide, biliverdin, and free iron [76]. Evidence suggests that the administration of CAP in a murine model reduces the production of pro-inflammatory cytokines and oxidative stress factors in the hypothalamic paraventricular nucleus via activation of the Nrf2/HMOX1 pathway [77]. Furthermore, a CAP diet (0.0015%) in a diabetic type 2 murine model has been shown to inhibit ferroptosis by activating TRPV1 and the Nrf2/HMOX1 signaling pathway. Nonetheless, there is a notable lack of experimental studies directly evaluating HMOX1 expression in neurodegenerative models treated with CAP [78]. NGF is a gene that encodes a neuropeptide, the nerve growth factor, which is produced during injury or inflammation and increases thermal sensitivity [79]. It has been implicated in the growth and survival of mammalian peripheral sensory and sympathetic nerve cells [80]. A study conducted on neonatal murine neurons demonstrated that NGF interacts with CAP, thereby enhancing the Ca²⁺ influx mediated by TRPV1 [81]. Furthermore, this neuropeptide has been proposed (in a narrative review) as a therapeutic target in AD to enhance cholinergic neuronal survival and plasticity in the brain [82].

Finally, we validated the molecular docking results for SIRT1, ESR1, HMOX1, and NGF using molecular dynamics simulations. Notably, SIRT1 and ESR1 underwent structural adjustments, resulting in increased compaction and stability, as evidenced by reduced Rg and SASA. This was further corroborated by the RMSD evolving towards constant values. Conversely, the HMOX1 and NGF complexes exhibited sustained structural stability throughout the simulation with minimal fluctuations, indicating a more passive interaction with CAP. As part of the molecular dynamics analysis, MM/PBSA data indicate that the ESR1-CAP complex exhibits the highest binding affinity, suggesting a remarkably stable interaction between ESR1 and CAP. As previously indicated, CAP can directly interact with ESR1 and facilitate estradiol binding to these receptors. In a murine model, 17β-estradiol has been identified as a highly potent agent against neuroinflammation, synaptic loss, and neurodegeneration [83]. One study found that elevated estrogen levels throughout life confer protection against PD [84]. Additionally, research conducted by Adams and Kumar [85] in 2013 documented the beneficial effects of 17β-estradiol in a male patient with PD who experienced motor fluctuations and dyskinesia. In this context, CAP could be considered as a possible adjunct to hormonal therapies for the treatment of AD and PD. However, further studies are warranted.

The energy differences among the NGF-CAP, HMOX1-CAP, and SIRT1-CAP complexes were minimal, approximately 0.30 kcal/mol. This suggests that, aside from ESR1, these systems exhibit comparable binding affinities and are equally favorable. In addition, the intermolecular van der Waals and electrostatic interactions are favorable for binding [86]. However, in this study, the high values of ΔE_vdw reflect that the term ΔE_vdw constitutes the main stabilizing force in the complexes, exceeding the electrostatic contribution ΔE_ele. These findings propose the structural viability of the formed complexes and suggest that CAP can interact favorably within the protein environment. Moreover, these findings suggest that although CAP can activate some of these targets in a TRPV1-dependent manner, it may also act through direct interactions.

In summary, the in silico results position CAP as a compound capable of interacting with different targets involved in AD and PD (independently of positive and negative effects). This fact opens future directions for research aimed at standardizing the administration of safe doses, exploring the role CAP may play in conjunction with pharmacological therapies used for the treatment and management of AD and PD, and understanding at which stages of these neurodegenerative disorders it may have a significant effect. Additionally, it is important to note that our findings were derived exclusively from network pharmacology, molecular docking, and molecular dynamics, indicating interactions between CAP and targets involved in AD and PD. Thus, these in silico predictions necessitate systematic experimental validation. We propose a rigorous evaluation of neuroprotective activity in AD and PD models using iPSC-derived or SH-SY5Y neurons that simultaneously explore the targets analyzed in this work. Subsequently, a thorough evaluation of its safety and efficacy should be conducted in animal models. Toxicological and pharmacokinetic analyses should be performed to corroborate its translational potential. Furthermore, we recognize the limitations inherent in the in silico approach, including reliance on databases with limited coverage, the inflexibility of docking models, the limited duration of molecular dynamics simulations (100 ns), and inaccurate entropy estimates from the MM/PBSA method.