Neuroprotective effects of essential oils in animal models of Alzheimer’s and Parkinson’s disease: a systematic review

Protocol and registration

This systematic review was conducted under the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines and was registered in a designated database (10.17605/OSF.IO/6MF78).

Information sources and search

Search descriptors were selected based on the Population/Problem, Interest, and Context (PICo) strategy. This review focused on experimental in vivo studies examining the neuroprotective role of essential oils (P) in neurodegenerative diseases (Co).

The databases searched included PubMed (US National Library of Medicine National Institutes of Health), Virtual Health Library, and Web of Science, utilizing the MeSH descriptors “essential oils,” “neuroprotection,” “Parkinson’s Disease,” “Alzheimer’s Disease,” “Huntington Disease,” and “Amyotrophic Lateral Sclerosis.” The search strategy was tailored to each database following the PICo inclusion criteria (detailed in the Supplemental Material). Additionally, search alerts were established to identify new studies.

Study selection and data collection

Studies published in English without date restrictions were considered. The inclusion criteria were experimental in vivo studies investigating the potential neuroprotective effects of essential oils on neurodegenerative diseases. Exclusions were made for reviews, case reports, clinical studies, and in vivo experimental studies not focusing on essential oils. Studies using plant extracts or derivatives instead of essential oils were also excluded. Below, are the eligibility criteria (inclusion/exclusion) in detail.

Risk of bias assessment

The risk of bias in individual studies was evaluated using the Systematic Review Center for Laboratory Animal Experimentation (SYRCLE) tool, adapted from the Cochrane Risk of Bias (RoB) tool for animal intervention studies (Hooijmans et al., 2014). This assessment was carried out by two independent reviewers (AM and TF) across ten domains to determine methodological quality: Sequence Generation, Baseline Characteristics, Allocation Concealment, Random Housing, Blinding (Performance Bias), Random Outcome Assessment, Blinding (Detection Bias), Incomplete Outcome Data, Selective Outcome Reporting, and Other Sources of Bias.

In the evaluation of methodological soundness, articles predominantly receiving “no” responses were deemed methodologically robust and indicative of a low risk of bias. Conversely, articles with a preponderance of “yes” responses were categorized as having a high risk of bias. Instances where the risk of bias could not be ascertained due to insufficient information were classified as having an “unclear risk.

To further elucidate the risk of bias, reviewers addressed additional questions:

• 1.

  Is there potential for the results to be biased?

• 2.

  Are there confounding factors that could obscure the interpretation of the results?

• 3.

  Is there a possibility that the study results are attributable to chance?

Study selection

During the initial database screening, a total of seventy-seven articles were identified. Subsequent application of the Rayyan selection tool facilitated the exclusion of 30 duplicates and 23 articles that did not meet the eligibility criteria. This refinement process resulted in twenty-four articles being selected for full-text review. Of these, seven articles were excluded due to their reliance on in vivo experimental designs not aligned with our criteria, and one article was excluded because it utilized Drosophila as the animal model, which was outside the scope of our review. Additionally, three articles were excluded for not specifically addressing the neuroprotective effects of essential oils. Consequently, thirteen articles were deemed eligible for qualitative assessment, as illustrated in Fig. 1.

Study characteristics

The included studies exhibited variability in several key areas, including the type of essential oil (EO) investigated (Fig. 2). In terms of EO extraction methods, hydrodistillation was the most common, used in eight articles, accounting for 66.66% of the studies (Postu et al., 2019; Lee, Chu & Chiang, 2021; Gao et al., 2019; Beserra-Filho et al., 2019; Postu et al., 2022; de Campos et al., 2023; de Lucena et al., 2020; Beserra-Filho et al., 2022). Steam distillation was reported in one article, representing 7.7% (Banji, Banji & Srinivas, 2021), and supercritical carbon dioxide extraction was also noted in one article, comprising 7.7% (Xu et al., 2023). Three articles did not specify the extraction method used (Liu et al., 2020; Jayaraj et al., 2022; Aykac et al., 2022).

The mode of EO administration varied across the studies: oral administration was documented in five articles (38.4%) (Banji, Banji & Srinivas, 2021; Jayaraj et al., 2022; Beserra-Filho et al., 2019; Aykac et al., 2022; Beserra-Filho et al., 2022), gavage in three articles (23.1%) (Lee, Chu & Chiang, 2021; Xu et al., 2023; de Lucena et al., 2020), inhalation in three articles (23.1%) (Postu et al., 2019; Liu et al., 2020; Postu et al., 2022), intraperitoneal administration in one article (7.7%) (de Campos et al., 2023), and intragastric administration in one article (7.7%) (Gao et al., 2019).

The research predominantly focused on AD and PD, representing 69.3% and 30.7% of the studies, respectively. AD induction methods included intracerebroventricular and intrahippocampal administration of β-amyloid (Postu et al., 2019; Lee, Chu & Chiang, 2021; Gao et al., 2019; Postu et al., 2022), intraperitoneal administration of aluminum chloride (Banji, Banji & Srinivas, 2021), intraperitoneal administration of scopolamine (Aykac et al., 2022; de Campos et al., 2023), and the use of transgenic mice as a disease model (Xu et al., 2023; Liu et al., 2020). PD models involved intraperitoneal administration of rotenone (Jayaraj et al., 2022), subcutaneous administration of reserpine (Beserra-Filho et al., 2022, 2019), and intracerebral administration of 6-hydroxydopamine (de Lucena et al., 2020).

Regarding animal models, the Wistar rat lineage was the most frequently used, present in six studies (46.15%) (Postu et al., 2019; Jayaraj et al., 2022; Postu et al., 2022; de Campos et al., 2023; Aykac et al., 2022). Swiss mice were utilized in two studies (15.4%) (Beserra-Filho et al., 2019, 2022), and transgenic mice were also used in two studies (15.4%) (Liu et al., 2020; Xu et al., 2023). Additionally, one study each employed Swiss albino, Kunming, and C57BL/6J lineages (Banji, Banji & Srinivas, 2021; Gao et al., 2019; Lee, Chu & Chiang, 2021).

Risk of bias within studies

The evaluation of methodological quality across ten domains revealed that most studies were classified as having a “low risk of bias,” indicating robust study methodologies among the included articles (Fig. 3). Notably, no studies were identified as possessing a high risk of bias, underscoring the methodological soundness of the research under review.

Specifically, the domains of “baseline characteristics,” “random housing,” “random outcome assessment,” “selective reporting,” and “other sources of bias” were deemed 100% adequate, reflecting optimal methodological practices (Fig. 4).

In the domain of random outcome assessment, 92.3% of the studies demonstrated methodological efficiency. However, one study presented ambiguity in methodological details, resulting in an “unclear risk” classification (Postu et al., 2019). Conversely, the domains of random sequence generation, allocation concealment, blinding of participants and personnel, and incomplete outcome data were categorized as having an “unclear risk of bias,” indicating areas where methodological clarity and rigor could be improved.

Results of individual studies and synthesis

In this systematic review, nine out of thirteen articles employed AD models to explore the neuroprotective role of essential oils (EOs) through their antioxidant effects, synaptic plasticity, and enhancement of learning and memory formation (Postu et al., 2019; Liu et al., 2020; Lee, Chu & Chiang, 2021; Gao et al., 2019; Xu et al., 2023; Postu et al., 2022; Lee, Chu & Chiang, 2021; de Campos et al., 2023; Aykac et al., 2022). Detailed results of these individual studies are presented in Table 1.

Postu et al. (2019, 2022) investigated the EO of P. halepensis Mill and demonstrated its nootropic and neuroprotective effects, evidenced by reduced memory impairments and oxidative damage in the hippocampus of rats. Citrus limon EO was also found to restore learning and memory in a murine model. Studies by Gao et al. (2019) and Xu et al. (2023) utilized EOs from the genus Acorus (Acorus gramineus and Acorus tatarinowii Schott) to treat cognitive dysfunction in AD.

Aiming to develop functional drugs for AD prophylaxis, EOs such as Litsea cubeba, Aniba canelilla, and Origanum onites were shown to have a prophylactic effect on memory deficits (Liu et al., 2020; Lee, Chu & Chiang, 2021; de Campos et al., 2023; Aykac et al., 2022). Banji, Banji & Srinivas (2021) demonstrated the therapeutic potential of combining turmeric extract and EO (Curcuma longa L., Zingiberaceae) in an aluminum-induced neurotoxic animal model.

In studies focusing on PD as the neurodegenerative disease model, four out of thirteen articles investigated the biological properties of EOs to reduce medication side effects and disease progression (Jayaraj et al., 2022; Beserra-Filho et al., 2019, 2022; de Lucena et al., 2020).

Jayaraj et al. (2022) demonstrated the neuroprotective role of citronellol against rotenone-induced neurodegeneration, with its antioxidant, anti-inflammatory, anti-apoptotic, and autophagy-modulating properties protecting dopaminergic neurons. Similar properties were observed with L-Linalool in hemiparkinsonian rats (de Lucena et al., 2020). EOs from Eplingiella fruticosa and Lippia grata were analyzed for their potential neuroprotective effects (Beserra-Filho et al., 2019, 2022). In these specific studies, the essential oils were combined with β-cyclodextrin to improve their solubility and stability, which enhanced the neuroprotective effects by increasing bioavailability and absorption, as highlighted in their research.

Anti-inflammatory effects

Neuroinflammation is a key driver of neuronal degeneration, as the prolonged activation of glial cells leads to the excessive release of pro-inflammatory cytokines and neurotoxic factors, exacerbating tissue damage. The activation of the TLR4/NF-κB pathway is one of the main contributors to this process, as it amplifies the expression of TNF-α, IL-1β, and IL-6, cytokines that contribute to excitotoxicity, increased oxidative stress, and neuronal death.

Given this scenario, essential oils have demonstrated multiple strategies for modulating neuroinflammation. The most prominent effect is their ability to block the activation of the TLR4/NF-κB pathway, thereby preventing the escalation of inflammatory responses. For example, Origanum onites reduced IL-1β immunoreactivity, an effect that can be attributed to inhibition of microglial activation and suppression of inflammatory gene transcription mediated by NF-κB (Aykac et al., 2022). Similarly, citronellol demonstrated suppression of TNF-α, IL-1β, and IL-6 in a Parkinson’s disease model, indicating a direct action in disrupting the NF-κB-mediated inflammatory cascade (Jayaraj et al., 2022).

Beyond their effects on the TLR4/NF-κB pathway, essential oils also influence astrocyte reactivity, a critical factor in sustained neuroinflammation. When activated, astrocytes increase their expression of glial fibrillary acidic protein (GFAP), altering their homeostatic functions and contributing to the inflammatory microenvironment. The essential oil of Aniba canelilla significantly reduced GFAP expression, suggesting that its anti-inflammatory action extends beyond microglia by also modulating astrocytic reactivity (de Campos et al., 2023). This modulation is particularly relevant, as it indicates that essential oils may prevent excessive recruitment of reactive astrocytes, thereby avoiding excessive glutamate release and other neurotoxic molecules.

Another important finding was the inhibition of the NLRP3 inflammasome, an intracellular protein complex that plays a central role in chronic inflammatory activation. NLRP3 activation is frequently associated with increased active caspase-1 and IL-1β production, events that intensify inflammation and oxidative stress. The essential oil of Acorus tatarinowii was found to suppress NLRP3 activation and reduce IL-1β levels in experimental models, indicating a potential to halt the perpetuation of the inflammatory response (Xu et al., 2023). Thus, essential oils not only reduce inflammatory markers but also target key cellular pathways, interrupting pro-inflammatory signals before they can amplify neuronal damage (Fig. 6). This integrated effect makes them promising candidates for therapeutic strategies aimed at disrupting the chronic inflammatory cycle characteristic of neurodegenerative diseases.

Antioxidant properties

Oxidative stress plays a crucial role in neurodegeneration as it induces lipid, protein, and nucleic acid damage, disrupting neuronal homeostasis. Lipid peroxidation, mediated by free radicals and reactive oxygen species (ROS), leads to the formation of toxic aldehydes such as malondialdehyde (MDA), a well-established biomarker of oxidative damage. Treatment with essential oils has shown the ability to reduce MDA levels in experimental neurodegeneration models. For instance, the essential oil of Pinus halepensis significantly reduced MDA levels in the hippocampus of rats treated with Aβ1-42, a widely used experimental model of Alzheimer’s disease, as determined by thiobarbituric acid reactive substances (TBARS) assay (Postu et al., 2019). Similarly, Litsea cubeba treatment lowered MDA levels and tau hyperphosphorylation, both biomarkers of neurodegeneration (Lee, Chu & Chiang, 2021).

Beyond reducing lipid peroxidation biomarkers, essential oils exert antioxidant effects by regulating endogenous enzymes involved in oxidative stress defense. Superoxide dismutase (SOD) catalyzes the conversion of highly reactive superoxide anions into hydrogen peroxide, which is subsequently degraded by catalase (CAT) and glutathione peroxidase (GPX). The essential oil of Lippia grata significantly increased the activity of these enzymes, reducing the oxidative burden in the prefrontal cortex and hippocampus of experimental Parkinson’s models (Beserra-Filho et al., 2022). This effect was further corroborated by Eplingiella fruticosa, which increased SOD and CAT levels, mitigating oxidative damage induced by neurotoxins (Beserra-Filho et al., 2019). Regulation of reduced glutathione (GSH) levels was also observed, suggesting that essential oils may play a role in intracellular redox balance modulation. The essential oil of Aniba canelilla restored GSH levels in rats subjected to scopolamine-induced cognitive deficits, highlighting a potential protective mechanism against ROS-mediated neurotoxicity (de Campos et al., 2023).

Neurotrophic effects

The loss of synaptic connectivity is an early and critical event in neurodegenerative diseases, being strongly influenced by neurotrophic factors. Brain-derived neurotrophic factor (BDNF) and its receptor TrkB play an essential role in maintaining synaptic plasticity, neuronal survival, and circuit remodeling. The essential oil of Acorus gramineus significantly increased BDNF, TrkB, and NT3 levels (Fig. 6), promoting neurogenesis and improved cognitive performance in experimental Alzheimer’s models (Gao et al., 2019). Additionally, treatment with lemon essential oil led to increased expression of synaptic proteins such as PSD95 and synaptophysin, suggesting a direct effect on maintaining synaptic architecture and neuronal functionality (Liu et al., 2020).

Integrated neuroprotective outcomes

The analyzed studies provide compelling evidence that essential oils exert neuroprotective effects by modulating oxidative stress and neuroinflammation, while also promoting synaptic plasticity and neurogenesis. Figure 6 provides a schematic illustration depicting the multiple pathways through which essential oils exert their neuroprotective effects. The diagram highlights four main therapeutic mechanisms: antioxidant effects, anti-inflammatory actions, neurotrophic effects, and integrated neuroprotective outcomes. Among the bioactive compounds present in essential oils, limonene and linalool emerge as promising candidates in suppressing the TLR4/NF-κB pathway, a critical inflammatory axis in neurodegenerative disease pathophysiology. Studies demonstrate that both compounds reduce the activation of this pathway, blocking the transcription of pro-inflammatory genes and minimizing oxidative stress (Mohamed, Abduldaium & Younis, 2020; Kathem et al., 2024).

The synergistic action of limonene and linalool may amplify their neuroprotective effects by simultaneously modulating two interconnected processes: neuroinflammation and oxidative stress. While limonene has been shown to reduce TLR4 receptor activation, thereby minimizing the initial recruitment of inflammatory responses, linalool inhibits NF-κB nuclear translocation, preventing sustained expression of pro-inflammatory cytokines. Furthermore, both compounds enhance endogenous antioxidant enzyme activity, such as SOD, CAT, and GPX, suggesting a protective effect against oxidative damage exacerbated by chronic inflammation.

This convergence of effects not only suggests a complementary potential but also raises the hypothesis that the combination of limonene and linalool may be more effective than their isolated use in regulating neuroinflammation and redox homeostasis. Although this synergy has yet to be extensively explored in clinical trials, experimental findings indicate that interactions between these monoterpenes may optimize neuroprotection and slow neurodegeneration progression.

Future studies should investigate the bioavailability and pharmacokinetics of limonene and linalool combinations, as well as their effects in clinical models of Alzheimer’s and Parkinson’s disease. Additionally, further research is needed to elucidate the cellular mechanisms underlying their synergistic effects, including their impact on neuronal apoptosis pathways, glial activation, and neurotrophic factor regulation. A deeper understanding of these interactions could open new avenues for developing more effective and safer natural therapies for neurodegenerative diseases, potentially complementing or even replacing conventional pharmacological approaches.

Limitations of the studies and future research directions

While the reviewed studies provide promising evidence for the neuroprotective effects of EOs, several methodological and conceptual limitations must be considered. These limitations not only restrict the generalizability of the findings but also highlight critical research gaps that must be addressed to validate the safety and efficacy of EOs in the context of neurodegenerative diseases.

One of the primary limitations is the predominance of studies conducted in animal models, which may not accurately reflect human physiological responses due to differences in metabolism, immune function, and neuronal signaling. The absence of rigorous clinical trials prevents definitive conclusions regarding the bioavailability, toxicity, and efficacy of EOs in patients with Alzheimer’s and Parkinson’s disease. Therefore, well-designed, placebo-controlled clinical trials are essential to determine the feasibility of EOs as therapeutic strategies, assessing their effects on inflammatory, oxidative, and neurotrophic biomarkers in human populations.

Moreover, the small sample sizes in most studies limit statistical power and hinder the reproducibility of findings. Future studies should address this limitation by increasing sample sizes and employing robust statistical analyses to enhance the reliability of conclusions regarding the neuroprotective potential of EOs.

Another critical limitation is the underrepresentation of female animals in experimental models. Of the 13 studies analyzed, only two included female subjects, raising concerns about potential sex-based differences in oxidative stress regulation, neuroinflammation, and synaptic plasticity. Hormonal factors—particularly estrogen levels—can significantly influence neuroprotective mechanisms and may affect EO efficacy. Future research should ensure balanced sex representation to achieve a more comprehensive understanding of EO efficacy and to identify sex-specific therapeutic responses.

Our ability to assess certain confounding variables was also limited by inconsistent reporting across studies. Environmental factors (e.g., housing conditions, light/dark cycles, handling procedures), comprehensive health status assessments of animal models prior to experimentation, and discussions of potential placebo effects in behavioral tests were rarely detailed. This reporting gap reflects a broader challenge in preclinical neuroscience research that may impact the interpretation of results. Future studies should prioritize transparent reporting of these variables to enhance reproducibility and translational relevance.

Chemical composition heterogeneity across EO samples remains a major challenge. The concentration of bioactive constituents can vary substantially due to differences in plant origin, environmental conditions, and extraction methods, thereby complicating cross-study comparisons and replication. For instance, Citrus limon EO showed substantial variation in limonene content across studies, making it difficult to attribute observed effects to a specific compound. To address this, a systematic approach to EO standardization is essential—one that encompasses both compositional and functional dimensions and enables meaningful comparisons while acknowledging the complex nature of these natural products. To this end, we propose a five-tier framework to guide future EO standardization efforts in neurodegenerative disease research:

• Chemical fingerprinting through advanced chromatographic techniques (GC-MS, HPLC-MS), with mandatory reporting of at least 95% of identified constituents and their relative concentrations;

• Bioactive compound quantification focusing on well-established neuroprotective molecules (e.g., limonene, linalool, β-caryophyllene), with clearly defined acceptance ranges for therapeutic consistency (e.g., Citrus limon EO should contain 30–45% limonene);

• Biological standardization using in vitro assays to measure specific activities (e.g., antioxidant capacity via DPPH/FRAP assays, anti-inflammatory potential via NO inhibition in LPS-stimulated microglia), with established minimum potency thresholds;

• Documentation of production parameters, including plant source geographical origin, cultivation conditions, harvesting time, extraction method parameters (temperature, pressure, duration), and storage conditions;

• Batch-to-batch consistency verification through both chemical and biological standardization assays, including stability testing across the proposed shelf life of each formulation.

In addition to compositional variability, the studies reviewed exhibited substantial variation in EO dosage and administration routes, which can significantly influence absorption, distribution, and metabolism. The lack of well-established protocols for EO administration in neurodegenerative models limits cross-study comparability. Future research should investigate EO pharmacokinetics and establish optimal dosages and delivery routes to maximize neuroprotective outcomes.

Importantly, most studies focused on early stages of disease progression, which may not fully capture the therapeutic potential of EOs across the entire neurodegenerative continuum. This focus on early intervention, while valuable for preventive applications, creates a significant knowledge gap regarding EO efficacy in moderate and advanced disease stages. Future animal studies should adopt staged intervention protocols, initiating EO treatment at multiple predefined timepoints based on established behavioral or biochemical markers. For example, in transgenic models of Alzheimer’s disease, parallel cohorts could receive identical EO interventions starting at early, moderate, and late stages. Similarly, in toxin-induced Parkinson’s models, staggered treatment timing relative to neurotoxin administration could provide insights into stage-specific therapeutic effects. Although resource-intensive and methodologically challenging, such approaches would enhance the translational relevance of EO-based interventions, particularly given that clinical populations often present at various stages of disease progression, and treatment options for advanced neurodegeneration remain severely limited.

To facilitate clinical translation, we also propose specific methodological frameworks for human trials. Initial efficacy trials should employ observer-masked, crossover randomized designs, allowing participants to serve as their own controls, whereas larger confirmatory trials should adopt parallel-group designs with demographic stratification. Outcome measures should integrate objective biomarkers with validated subjective assessments at multiple timepoints (e.g., baseline, 4 weeks, 12 weeks) to assess both immediate and sustained effects. Long-term safety requires dedicated longitudinal studies lasting 6–12 months post-intervention, including comprehensive dermatological, hepatic, and immunological monitoring. Dose-response studies are essential for defining therapeutic windows for chronic administration.

Potential interactions with conventional medications must be systematically evaluated, particularly for drugs frequently co-administered with EOs and those metabolized via pathways influenced by terpenes. In vitro assessments of cytochrome P450 interactions should precede clinical trials using established probe substrates. Future clinical studies must also prioritize population diversity through stratified recruitment strategies that ensure representation across age groups, ethnicities, and genetic backgrounds. Special consideration should be given to vulnerable populations (e.g., pediatric, pregnant/lactating, immunocompromised), which may require population-specific dosing algorithms informed by personalized medicine approaches already established in other therapeutic areas.

In summary, future research on essential oils should be methodologically rigorous, with larger and more representative samples, standardized chemical and biological protocols, well-designed clinical trials, and innovative delivery systems. Addressing these limitations will be critical to establishing essential oils as viable therapeutic agents for neurodegenerative diseases.