Mapping Toxicological Trends: A Bibliometric Analysis of Organophosphate and Flavonoid Research in Animal Models
Article information
Abstract
Organophosphates (OPs) remain persistent environmental contaminants, while flavonoids are increasingly recognized for their protective roles against toxic insults. This study employed a structured bibliometric analysis to evaluate global research trends on OP toxicity and flavonoid interventions using six commonly used animal models: zebrafish, rat, mouse, fruit fly, frog, and nematode. Data were retrieved from PubMed (1990–2025), and analyzed through publication counts, network mapping, and compound annual growth rate (CAGR). Co-occurrence keyword networks and overlay visualizations were generated using VOSviewer. Results revealed the continued dominance of rodent models in both OP and flavonoid studies, with rat- and mouse-based literature exhibiting the highest total link strength. However, zebrafish showed the highest CAGR for OP-related publications (12.78%) and robust network development in flavonoid research, suggesting a rising preference for this model. Zebrafish-flavonoid clusters were thematically rich, with emerging keywords such as "Nrf2 signaling" and "blood-brain barrier," indicating expanding mechanistic inquiry. Caenorhabditis also demonstrated notable growth in flavonoid literature, underscoring the relevance of invertebrate models. This analysis highlights a growing shift toward scalable, ethical, and mechanistically diverse model organisms. These trends support zebrafish’s role as a bridge model between environmental toxicology and biomedical translation. Bibliometric approaches such as this can inform future research priorities, funding allocations, and policy-making by identifying underexplored models and evolving scientific paradigms.
Introduction
Organophosphate (OP) pesticides continue to play a pivotal role in global agriculture due to their high efficacy against a broad range of insect pests. However, this utility comes at a cost. The widespread and often unregulated use of OPs has led to severe environmental contamination and public health risks. Current estimates suggest that OP exposure is responsible for approximately 300,000 deaths annually, predominantly in low- and middle-income countries [1]. These figures underscore a pressing need to explore both the acute and chronic health consequences associated with OP exposure.
Historically, toxicological research has focused on the acute effects of OPs, particularly their irreversible inhibition of acetylcholinesterase (AChE), which leads to cholinergic overstimulation and a cascade of neuromuscular symptoms. However, recent studies have expanded this narrative by revealing OPs' involvement in long-term neurodevelopmental impairments, cognitive decline, and neurodegenerative processes [2, 3]. These findings have redirected toxicological inquiry toward deeper molecular mechanisms such as oxidative stress, neuroinflammation, and apoptosis [4, 5].
The physicochemical properties of OPs allow them to persist in various environmental compartments, including air, soil, water bodies, and biota. Numerous studies have detected OP residues in food products (e.g., fruits, vegetables, tea, honey), drinking water, and even ambient air, highlighting their global dispersal [6, 7, 8]. Their environmental mobility also facilitates long-range transport, resulting in contamination of remote and coastal ecosystems. In aquatic systems, OPs bioaccumulate and biomagnify in organisms at higher trophic levels, causing imbalances in biodiversity and disrupting ecological homeostasis [9-11].
Human exposure to OPs occurs through multiple routes: inhalation of contaminated air, ingestion of polluted food and water, and dermal absorption. These exposure routes are particularly significant for agricultural workers, children, and pregnant women, whose biological susceptibility or occupational exposure increases risk [12, 13]. Epidemiological evidence has consistently associated OP exposure with a range of adverse outcomes including developmental delays, hormonal disruption, and neurobehavioral abnormalities [7].
Despite growing awareness, the implementation of regulatory restrictions on OP use varies greatly between regions. High-income countries have enacted partial or complete bans, while lower-income countries continue to use OPs extensively due to economic constraints and weak enforcement mechanisms. Global frameworks such as the Stockholm Convention and guidelines from WHO and FAO exist to promote pesticide safety, yet their effectiveness depends heavily on national adoption and infrastructure [14-16].
Beyond regulation, remediation technologies have been developed to address OP contamination, including microbial degradation and phytoremediation. While promising in controlled settings, these methods often struggle with scalability, field deployment, and efficacy under diverse environmental conditions [17, 18]. This has spurred further research into biological strategies that could serve both therapeutic and preventive roles.
One such strategy involves the use of phytochemicals—especially flavonoids—for mitigating OP-induced toxicity. Flavonoids are plant-derived polyphenolic compounds known for their antioxidant, anti-inflammatory, and neuroprotective properties. Molecules such as quercetin, kaempferol, and curcumin have demonstrated efficacy in counteracting OP toxicity in various animal models by modulating key signaling pathways, including Nrf2/ARE, NF-κB, and PPAR-γ [19-22]. These compounds are of particular interest due to their non-synthetic origin, favorable safety profile, and potential dual role as therapeutic agents and environmental safeguards.
Despite growing preclinical evidence, there remains a lack of integrative analysis that maps the current state of research on OP toxicity and flavonoid interventions. Previous reviews have tackled these areas individually—addressing OP residues in the environment, mechanisms of toxicity, or the pharmacology of flavonoids—but few have explored their intersections within a unified framework. This gap presents an opportunity for bibliometric approaches to elucidate how these fields converge, evolve, and inform one another.
Bibliometric analysis enables the systematic study of scientific literature using statistical and network-based methods. Recent advances in bibliometric tools have enhanced our ability to analyze large datasets, detect thematic clusters, identify influential publications, and track emerging trends across time [23, 24]. In toxicology, this allows researchers to better understand which models, mechanisms, and therapeutic agents dominate the discourse and where critical knowledge gaps lie.
In this study, we conducted a bibliometric analysis of research involving OP toxicity and flavonoid interventions using six animal models commonly employed in toxicological studies: zebrafish, rat, mouse, fruit fly, frog, and nematode. This approach integrates data on publication counts, keyword networks, and compound annual growth rates to map intellectual structures and assess the translational relevance of experimental findings. By identifying conceptual trajectories and model-specific patterns, we aim to provide strategic insights that can guide future research, inform policy, and align with the broader goals of One Health and environmental justice.
Materials and Methods
Study Design
This study employed a structured bibliometric approach to examine the research landscape surrounding organophosphate (OP) toxicity and flavonoid-based interventions in experimental animal models. Designed to identify thematic clusters, mechanistic patterns, and evolving research trends, the methodology integrates data-driven mapping with critical toxicological reasoning. Special emphasis was placed on the translational and environmental health relevance of six widely used animal models, capturing the intersection of mechanistic toxicology, phytochemical therapeutics, and alternative model development.
Data Collection
Data were retrieved from the PubMed database, which offers a comprehensive repository of biomedical, pharmacological, and toxicological literature. Between November 25 and December 5, 2024, twelve structured search queries were executed, each combining the terms "organophosphate" or "flavonoid" with the scientific names of six animal models: Danio rerio (zebrafish), Rattus norvegicus (rat), Mus musculus (mouse), Drosophila melanogaster (fruit fly), Xenopus laevis (frog), and Caenorhabditis elegans (nematode). This approach was guided by the goal of delineating compound-specific publication trends and identifying model-specific trajectories. The search window was set between 1990 and 2025 to ensure consistency with prior toxicological timelines and to capture shifts in experimental focus over three decades. Two types of outputs were generated for each query: annual publication distribution (CSV format) and raw article metadata (TXT format).
Data Processing
Timeline data were first processed using Microsoft Excel 2019 to generate descriptive visualizations of publication trends categorized by compound type and model organism. This included longitudinal trend graphs and stacked bar charts comparing the research output for OP and flavonoid studies across all six models. An example of the resulting trend visualization is presented in the Results section (Figure 2), illustrating publication trajectories by model and compound class. These graphs were designed to provide an intuitive overview of topic momentum, not to serve as inferential statistical evidence.
Annual publication trend of OP (left) and Flavonoid (right) research in six experimental models (1990–2024).
In addition to descriptive trend plots, two quantitative indicators were calculated to enrich interpretation: Annual Growth Rate (AGR) and Compound Annual Growth Rate (CAGR). AGR represents the year-over-year percentage change in publication volume and was used to highlight short-term fluctuations. It was calculated using the formula:
where Pt denotes the publication count in year t and Pt-1 in year t - 1.
CAGR captures the long-term average growth from the first year with non-zero publications to the most recent year available. It was calculated as:
where P0 and Pn refer to the publication counts in the start and end years respectively, and n is the total number of years between them.
Keyword Mapping and Bibliometric Analysis
For keyword analysis and network visualization, co-occurrence mapping was performed using VOSviewer (version 1.6.19). Full counting mode was selected, and a minimum keyword occurrence threshold of five was applied uniformly across all datasets. This threshold ensured focus on terms of sufficient recurrence to form stable clusters. To avoid visual saturation and optimize clarity, a 1,000-keyword ceiling was implemented for each compound-model pair. This practice follows established bibliometric conventions and aligns with the performance recommendations described by van Eck and Waltman (2010), the developers of VOSviewer.
Animal Model Rationale
The rationale for selecting six animal models was informed by their prominence in toxicological literature, availability of experimental tools, and representation in regulatory or preclinical research. Only models with more than 100 PubMed-indexed publications were included to ensure network density sufficient for bibliometric clustering. Vertebrate models such as zebrafish, rat, mouse, and frog offer varying degrees of physiological and genomic similarity to humans, while invertebrate models like fruit fly and nematode contribute scalability, genetic tractability, and ethical favorability. These models were evaluated against multiple criteria—genetic homology, experimental scalability, toxicological relevance, and regulatory status—using an adapted synthesis of previous reviews [25, 26].
Table 1 summarizes the comparative strengths and limitations of each model organism. Zebrafish, for example, provide high-throughput capabilities and strong genetic conservation but diverge metabolically due to ectothermy and whole-genome duplication. Rodents dominate toxicology due to physiological similarity and abundant genetic tools, though they present higher costs and stricter ethical constraints. Amphibians and invertebrates offer unique developmental and cellular advantages, though with limitations in disease modeling and immune complexity.
Comparative advantages and disadvantages of commonly used animal models in studies on organophosphate toxicity and flavonoid intervention. Evaluation criteria include genetic similarity to humans, experimental tractability, toxicological relevance, and ethical considerations.
Workflow Overview
The overall bibliometric workflow is outlined in Figure 1, which illustrates each step from search term design to data extraction, preprocessing, co-occurrence mapping, cluster detection, and visual overlay generation. Each step adheres to established protocols in bibliometric research, adapted here for dual thematic foci: OP toxicity and flavonoid mitigation. The workflow was further designed to allow reproducibility and future adaptation across other compound families or exposure models.
Workflow for bibliometric analysis. The diagram outlines keyword design, search execution, data extraction, analysis, and visualization using VOSviewer.
Reproducibility and Analytical Validity
To ensure methodological transparency and replicability, all keyword combinations were predefined and consistently applied across datasets. A temporal filter from 1990 to 2025 was enforced on all searches. The keyword threshold (n = 5) and ceiling (1,000) were selected to standardize network construction and facilitate cross-model comparisons. These parameters have been validated in previous large-scale bibliometric studies [23, 24] and were calibrated here to accommodate the dual complexity of chemical class and model organism.
Results and Discussion
Publication Output by Animal Model and Compound Class
The bibliometric search retrieved 12 datasets corresponding to combinations of organophosphate (OP) and flavonoid topics with six experimental models. Overall, rodent models (rats and mice) remain the predominant subjects in OP-related and flavonoid-related publications. As shown in Table 2, rat-based studies accounted for 3,690 OP publications and 26,741 flavonoid articles, while mice yielded 2,181 and 24,440 respectively. In contrast, zebrafish studies on OPs were limited to 424 articles, but reached 612 publications in the flavonoid domain, showing a more recent but significant rise. These results reflect the historical dominance of mammalian systems in toxicological and pharmacological literature. However, the increasing visibility of zebrafish in flavonoid studies—particularly after 2015—is consistent with its emergence as a high-throughput, scalable, and ethically favorable model for screening phytochemicals [27].
Count of Articles from respective pairs of keywords (1990–2025)
Figure 2 illustrates longitudinal publication trends for organophosphate (OP) and flavonoid research across six experimental animal models from 1990 to 2024. For OP-related studies, rodent models—particularly rats and mice—maintained the highest output, peaking between 2005 and 2015 before gradually declining. In contrast, zebrafish research on flavonoids displayed a steep and consistent rise beginning in the mid-2010s, reflecting a shift toward non-mammalian, high-throughput models in phytochemical toxicology. Other models such as Drosophila, Xenopus, and Caenorhabditis remained low in volume across both compound categories, although slight upward trends are observable in recent years.
Network Dynamics: Conceptual Cohesion and Keyword Clustering
Co-occurrence mapping using VOSviewer revealed differing network structures by compound and model. Table 3 summarizes the number of clusters, inter-keyword links, and total link strength per keyword pair. For OPs, rat and mouse models generated the highest total link strengths (300,821 and 177,175 respectively), with zebrafish notably lower at 17,739. In flavonoid studies, however, zebrafish yielded 27,773 in total link strength, suggesting more robust conceptual cohesion than in its OP counterpart. These data reaffirm the entrenched status of rodent models while underscoring the conceptual maturation of zebrafish-flavonoid literature. The higher number of clusters in zebrafish-flavonoid mapping (n = 9) also suggests diversification in research focus, including neuroinflammation, oxidative stress, and behavioral endpoints [28].
Cluster count, total links, and link strength by model and compound class.
Temporal Mapping and Mechanistic Trends
Figures 3 and 4 present timeline keyword overlays for OP and flavonoid literature, respectively, broken down by animal model panels (A–F) corresponding to zebrafish, rat, mouse, fruit fly, frog, and nematode. In Figure 3, rodent-based OP studies exhibit keyword centrality around acetylcholinesterase inhibition, oxidative stress, and neurotoxicity. These terms cluster around the 2000–2015 period, as shown by the dominant color gradient in the overlay visualization. This pattern reflects historical mechanistic focus in OP toxicology, particularly within rat and mouse models [29]. Zebrafish overlays for OPs show later-emerging terms such as "developmental exposure" and "behavioral assay," aligning with zebrafish’s unique role in developmental toxicology [30].
Timeline overlay visualization of OP keyword co-occurrence maps across six animal models using VOSviewer. (A) Zebrafish (Danio rerio); (B) Rat (Rattus norvegicus); (C) Mouse (Mus musculus); (D) Fruit fly (Drosophila melanogaster); (E) Frog (Xenopus laevis); (F) Nematode (Caenorhabditis elegans). Color gradient reflects average publication year, with recent terms shown in yellow.
Timeline overlay visualization of flavonoid keyword co-occurrence maps across six animal models using VOSviewer. (A) Zebrafish (Danio rerio); (B) Rat (Rattus norvegicus); (C) Mouse (Mus musculus); (D) Fruit fly (Drosophila melanogaster); (E) Frog (Xenopus laevis); (F) Nematode (Caenorhabditis elegans). Color scale indicates average term occurrence by publication year.
In flavonoid studies (Figure 4), the zebrafish network features terms such as “antioxidants,” “Nrf2 signaling,” and “neuroprotection,” with more recent terms including “blood-brain barrier” and “inflammatory markers” emerging post-2018. These trends are consistent with a growing body of evidence supporting the neuroprotective effects of flavonoids like quercetin and kaempferol in zebrafish models. At low doses, quercetin has been shown to improve cognitive and social behaviors, reduce oxidative stress, and protect dopaminergic neurons, while also enhancing antioxidant enzyme activity and glutathione levels [31-33]. Furthermore, quercetin-loaded nanoparticles may restore blood-brain barrier (BBB) integrity and mitigate neuroinflammation, positioning flavonoids as promising candidates for environmentally relevant CNS protection [33, 34].
Publication Momentum and Model Emergence
To quantify research momentum, compound annual growth rate (CAGR) was calculated for each compound-model combination (Figure 5). Zebrafish showed the highest CAGR in OP-related studies (12.78%), while rat and Xenopus models displayed flat or negative growth. For flavonoid research, Caenorhabditis elegans and zebrafish recorded the strongest CAGRs at 17.3% and 16.3%, respectively.
CAGR (%) of organophosphate-related (left) and flavonoid-related (right) publications by animal model (1990–2024).
These trends suggest a growing recognition of zebrafish and C. elegans as scalable and biologically relevant models in toxicology and pharmacology. Zebrafish have been widely used to evaluate flavonoid-mediated protection against oxidative stress, neuroinflammation, and neurodegeneration, supported by evidence of improved behavioral and molecular outcomes following exposure to compounds such as agathisflavone, rutin, and apigenin [35-39]. Their utility in modeling age-related cognitive impairment and antioxidant signaling pathways (e.g., Nrf2, SIRT1) further reinforces their translational potential.
Likewise, C. elegans has enabled high-throughput investigations of flavonoid-induced longevity and stress resilience through well-conserved pathways such as DAF-16/FOXO and AMPK [40]. The increasing bibliometric footprint of these models reflects a broader shift toward alternative testing frameworks that are cost-efficient, ethically aligned, and mechanistically insightful.
Translational Implications: Zebrafish as a Bridge Model
Zebrafish’s alignment with key human-relevant pathways, notably Nrf2/ARE and blood-brain barrier modulation, underscores its translational potential. Its external fertilization, transparent embryos, and scalability suit both mechanistic and screening studies. When considered alongside the rapid publication growth identified in previous section, zebrafish clearly exemplifies a rising research investment toward ecologically compatible and mechanistically rich toxicological models. Despite its promise, the underutilization in OP toxicology points to a gap that, if addressed, could catalyze methodological innovation and regulatory impact [30]. The current toxicological canon may benefit from recalibration, incorporating zebrafish not as a supplementary model but as a frontline screening organism capable of bridging environmental and biomedical toxicology.
Environmental and Biomedical Integration
The integration of flavonoid research into the broader context of OP toxicology offers a promising avenue for preventive toxicology. Across model systems, mechanisms such as oxidative stress, apoptosis, and neuroinflammation consistently emerge as convergent toxicological targets. These shared pathways support the development of phytochemical-based mitigation strategies and align with the principles of One Health, which call for environmentally sustainable and ethically responsible alternatives to synthetic antidotes. Zebrafish and Caenorhabditis elegans, in particular, are gaining attention for their ability to bridge environmental and biomedical perspectives. Zebrafish models allow in vivo visualization of neurodevelopmental processes and behavioral endpoints, while C. elegans enables high-throughput, gene-targeted screening of aging and oxidative stress pathways [38-40]. Although Xenopus and Caenorhabditis remain underrepresented in overall publication volume, their unique physiological features—such as amphibian endocrine responsiveness and nematode genetic tractability—position them as valuable complements to zebrafish in tiered toxicity assessments addressing both ecological and translational endpoints.
Data Expansion for Strategic Insight
To advance bibliometric understanding, future studies should incorporate citation analysis to identify pivotal publications that have shaped the field. Mapping journal clusters may reveal thematic silos or interdisciplinary gaps that hinder knowledge transfer. Additionally, geographic and funding-source analyses could uncover structural inequalities in toxicological research capacity across regions. Beyond structural mapping, aligning specific toxicological endpoints—such as mitochondrial dysfunction, neurogenesis, or blood-brain barrier integrity—with model organisms through network or Sankey diagram visualizations would deepen insight into mechanistic distributions across systems. These data-driven expansions would enhance the bibliometric framework’s utility for regulatory science, translational medicine, and global environmental health policy.
Conclusions
This study offers a focused bibliometric assessment of organophosphate (OP) toxicity and flavonoid intervention research across six experimental models. While rats and mice remain dominant, zebrafish demonstrated the fastest growth, particularly in flavonoid literature, and showed the highest compound annual growth rate (CAGR) in OP-related studies.
Zebrafish-related flavonoid research formed conceptually cohesive clusters around neuroinflammation, oxidative stress, and neuroprotection, highlighting its value as a translational and ethical model. Caenorhabditis also emerged as a growing invertebrate model in flavonoid studies, emphasizing its relevance in mechanism-focused screening. By combining bibliometric data with toxicological insight, this study reveals evolving model preferences and conceptual shifts in the field. It informs regulatory science and One Health approaches, while highlighting zebrafish as a promising bridge between environmental and biomedical toxicology. Further studies may extend this framework to include toxicological endpoints, citation dynamics, and geographic disparities.
Notes
Acknowledgement
The authors acknowledge the institutional support and access to bibliometric databases provided by the Universitas Brawijaya Central Library.
Conflict of interest
The authors declare no conflict of interest.
CRediT author statement
D.S.C. and K.M.K. designed the study and performed the data analysis. D.Y.P. and I.N. contributed to data collection and figure preparation. R.A.M.A. assisted in literature synthesis and manuscript formatting. W.A.C. supervised the project, revised the manuscript critically, and is the corresponding author. All authors read and approved the final manuscript.
Supplementary Material
Supplementary material includes Scopus-based keyword co-occurrence network metrics by model and compound class (Table S1) and Scopus-based timeline overlay visualizations for organophosphate and flavonoid keyword co-occurrence maps across six animal models (Figures S1 and S2). This material is available online at www.eaht.org.