Regulatory trends of organophosphate and pyrethroid pesticides in cannabis and applications of the Comparative Toxicogenomics Database and Caenorhabditis elegans
School of Mathematical and Natural Sciences, Arizona State University, Glendale, AZ 85306, United States
ASU-Banner Neurodegenerative Disease Research Center, Arizona State University, Tempe, AZ 85281, United States
School of Mathematical and Natural Sciences, Arizona State University, Glendale, AZ 85306, United States
School of Mathematical and Natural Sciences, Arizona State University, Glendale, AZ 85306, United States
ASU-Banner Neurodegenerative Disease Research Center, Arizona State University, Tempe, AZ 85281, United States
College of Health Solutions, Arizona State University, Phoenix, AZ 85004, United States
School of Mathematical and Natural Sciences, Arizona State University, Glendale, AZ 85306, United States
School of Mathematical and Natural Sciences, Arizona State University, Glendale, AZ 85306, United States
Department of Translational Neuroscience, Michigan State University, Grand Rapids, MI 49503, United States
Francis King Carey School of Laws, University of Maryland, Baltimore, MD 21201, United States
School of Health Sciences, Purdue University, West Lafayette, IN 47907, United States
Purdue Institute for Integrative Neuroscience, Purdue University, West Lafayette, IN 47907, United States
School of Mathematical and Natural Sciences, Arizona State University, Glendale, AZ 85306, United States
ASU-Banner Neurodegenerative Disease Research Center, Arizona State University, Tempe, AZ 85281, United States
College of Health Solutions, Arizona State University, Phoenix, AZ 85004, United States
Abstract
Organophosphate and pyrethroid pesticides are common contaminants in cannabis. Due to the status of cannabis as an illicit Schedule I substance at the federal level, there are no unified national guidelines in the United States to mitigate the health risk of pesticide exposure in cannabis. Here, we examined the change in the state-level regulations of organophosphate and pyrethroid pesticides in cannabis. The medians of pyrethroid and organophosphate pesticides specified by each state-level jurisdiction increased from zero pesticide in 2019 to 4.5 pyrethroid and 7 organophosphate pesticides in 2023, respectively. Next, we evaluated the potential connections between pyrethroids, organophosphates, cannabinoids, and Parkinson’s disease using the Comparative Toxicogenomics Database (CTD). Eleven pyrethroids, 30 organophosphates, and 14 cannabinoids were associated with 95 genes to form 3,237 inferred and curated Chemical-Gene-Phenotype-Disease tetramers. Using a behavioral repulsion assay with the whole organism model Caenorhabditis elegans, we examined the effect of cannabinoids and insecticides on depleting dopamine synthesis. Exposure to chlorpyrifos and permethrin, but not Δ9-tetrahydrocannabinol (THC) and cannabidiol (CBD), results in dose-dependent effects on 1-nonanol repulsive behaviors in C. elegans, indicating dopaminergic neurotoxicity (P < 0.01). Dose-dependent effects of chlorpyrifos are different in the presence of Δ9-THC and CBD (P < 0.001). As a proof of concept, this study demonstrated how to use new approach methodologies such as C. elegans and the CTD to inform further testing and pesticide regulations in cannabis by chemical class.
Untitled section
Keywords: cannabis, new approach methodology, systems biology, regulatory policy, pesticide, Caenorhabditis elegans
Article notes
Untitled section
Collection date 2025 Apr.
In recent years, public interest in the medical value of cannabis has grown significantly. Thirty-eight states and Washington, D.C. have now legalized the medical use of cannabis including use for alleviating pain, symptoms related to neurological dysfunction, and psychological conditions such as PTSD (Pinkhasova et al. 2021; NCSL 2024). Parkinson’s disease, the second most common neurodegenerative disorder, is currently listed as a qualifying condition by medical cannabis programs in 16 states (Griffith et al. 2024). Many individuals with Parkinson’s disease use cannabis as a supplement to alleviate the nonmotor symptoms of Parkinson’s disease, including pain, anxiety, and sleep disorders (Figura et al. 2022). In 2 recent large-scale surveys, 25% to 40% of individuals with Parkinson’s disease reported using cannabis as an alternative therapy (Kindred et al. 2017; Feeney et al. 2021). Yet, cannabis and cannabinoid products (except Epidiolex) have not undergone the U.S. Food and Drug Administration approval process and are not under the same manufacturing and quality controls as prescription drugs (Pruyn et al. 2022). The lack of quality control and safety standards for cannabis-based products can lead to adverse health outcomes in medical use due to contaminant exposure (Dryburgh et al. 2018; Jameson et al. 2022).
Despite recent interest in medical cannabis in neurological diseases, cannabis product safety—especially the risk of unregulated pesticides—remains a research gap that needs to be addressed (Pinkhasova et al. 2021; Schauer et al. 2023). Due to the Schedule I status of cannabis, the regulation of pesticidal residues in cannabis and cannabinoid products is mostly absent at the federal level (Jameson et al. 2022). The U.S. Environmental Protection Agency (U.S. EPA) has not issued any guidance or tolerance level on pesticide use in cannabis. The U.S. Department of Agriculture is not monitoring the pesticide residue level in cannabis as it is for other agricultural commodities. A recent study by the Los Angeles Times found 25 of 42 legal cannabis products containing pesticides that were above the levels allowed for cannabis in California or, for unregulated pesticides, levels above standards for tobacco products (St John and Halperin 2024). Organophosphate and pyrethroid pesticides were commonly found contaminants in cannabis (Jameson et al. 2022; St John and Halperin 2024). They are also known human neurotoxicants with potential links to Parkinson’s disease (Narayan et al. 2013; Wang et al. 2014; Furlong et al. 2015). Yet, little is known about how cannabis-borne exposure to organophosphate and pyrethroid pesticides may affect Parkinson’s disease patients, whereas exposure to organophosphate pesticides is associated with accelerated symptom progression in individuals with Parkinson’s disease (Li et al. 2023).
New approach methodologies (NAMs)—including the Comparative Toxicogenomics Database (CTD) and the whole organism model Caenorhabditis elegans—provide novel tools for predicting potential health hazards of cannabis-borne pesticide exposure. These predictive tools are rapid, inexpensive methods to assess exposure and toxicity (Kavlock et al. 2018; Serafini et al. 2024). They are useful in understanding mixture toxicity which can be higher than the sum of its components (Kortenkamp and Faust 2018). For example, the CTD can identify the potential mechanistic connections between environmental exposure and adverse health outcomes (Davis et al. 2021). It highlighted the multiple disturbed biological pathways of air-borne pollutants (Rager et al. 2011) and male reproductive toxicants (Davis et al. 2023). C. elegans is well-studied for their behavior responses and genetic similarity to humans (Leung et al. 2008; Wittkowski et al. 2019). It has been used in multiple mixture toxicity studies, including heavy metal mixtures (Tang et al. 2019), flavor chemicals (Lu et al. 2021), and pesticides (Wang et al. 2021).
In this study, we examined the change in the regulations of organophosphate and pyrethroid pesticides in cannabis in the United States by comparing the state-level regulations from 2019 to 2023. As a proof of concept, we used a combination of NAMs to examine a hypothetical exposure scenario where individuals with Parkinson’s disease are exposed to cannabis contaminated by organophosphate and pyrethroid pesticides. We first examined the gene-level connections of pesticide-cannabinoid mixtures by data-mining the CTD on organophosphate pesticides, pyrethroid pesticides, and cannabinoids. Next, we used C. elegans behavioral assays to examine the effect of chlorpyrifos, permethrin, Δ-9 tetrahydrocannabinol (THC), and cannabidiol (CBD) in causing dopaminergic neurotoxicity. Finally, we characterized the dose–response relationship of chlorpyrifos in the presence of Δ-9 THC and CBD.
Materials and methods
Reviewing state-level regulations of organophosphate and pyrethroid pesticides in cannabis
We compared the pesticide regulations of legalized cannabis in the United States between 2019 and 2023. We first reviewed the pesticide regulations of all legalized states in 2019 as summarized by the Arizona Department of Health Services (AZDHS 2019). Next, we searched for the pesticide regulatory documents on the government (.gov) websites of all legalized states and Washington D.C. in 2023 as described by Jameson et al. (2022). The document collection began on April 27 and ended on June 6, 2023. We identified all regulated organophosphate and pyrethroid pesticides and the regulatory action levels for each pesticide in 2019 and 2023. If a jurisdiction provided different action levels for different product categories (e.g. flower, inhalable, edible, etc.), only the levels for flower or inhalable were collected. Any jurisdictions that provided no specified regulatory action levels were excluded from further analysis. Additionally, we excluded any jurisdictions that used the full pesticide list of Title 40 of the Code of Federal Regulations Part 180 (C.F.R. 180): Tolerances and Exemptions for Pesticide Chemical Residues in Food (U.S. Environmental Protection Agency (U.S. EPA) 2024) as regulatory action levels, because the list contained some 400 pesticides and was not relevant to cannabis production (Jameson et al. 2022).
Analyzing data from the CTD
We evaluated the potential mechanistic connections between organophosphate and pyrethroid pesticides, different cannabinoids, and Parkinson’s disease using the CTD (data release: November 30, 2023, revision 17204). We used the “CTD Tetramer” tool in CTDBase.org and collected the Chemical-Gene-Phenotype-Disease (CGPD) tetramers as described by Davis et al. (2023). The tetramers were generated with (i) all organophosphate pesticides, pyrethroid pesticides, and cannabinoids available in the CTD and (ii) “Parkinson disease” and all descendent terms (e.g. “Parkinsonian disorders,” “Parkinson disease, late-onset,” and “Parkinsonism-dystonia, infantile”). We also developed a new methodology based on curated CGPD tetramers to identify broader mechanistic connections using the inferred chemical–disease associations. Three independently curated CTD datasets—inferred chemical–disease associations with specified inference genes, chemical–phenotype associations, and gene–GO annotations—were used to build sets of computational constructed information blocks (i.e. inferred CGPD tetramers) to connect a chemical, a gene, a phenotype, and Parkinson’s disease. Each inferred CGPD tetramer represented a hypothetical chemical-to-disease connection that met all 3 lines of evidence, whereas the presence of both an inferred and a curated CGPD tetramer indicated more weight (i.e. 4 lines) of evidence that supported the chemical-to-disease connection (Davis et al. 2023).
Data visualization
We counted the regulated organophosphate and pyrethroid pesticides in each US jurisdiction in 2019 and 2023 as previous described (Jameson et al. 2022). The distribution of the organophosphate and pyrethroid pesticide counts was visualized using overlaid histograms in RStudios (ggplot2 version 3.4.2, RStudios version 2023.06.0 + 421, R version 4.3.0). The action levels for the regulated organophosphates and pyrethroids were collected, and each pesticide was ranked by the total action levels listed for 2019 and 2023 regulations. The pesticides were ranked by the number of regulating jurisdictions and their action levels were visualized on a log base 10 scale with box plots and a strip plot overlay (ggplot2 version 3.4.2, RStudios version 2023.06.0 + 421, R version 4.3.0).
We used Gephi—a network analysis and visualization application (version 0.10.1; Bastian et al. 2009)—to assess the mechanistic relationships between cannabinoids, each pesticide class, and gene. We created 2 lists of chemical-to-gene connections (i.e. edges), one for cannabinoids and organophosphate pesticides and the other for cannabinoids and pyrethroid pesticides. Each chemical-to-gene connection on the lists was weighed by the number of tetramers (i.e. a weighted edge list). The lists were passed into Gephi to create 2 bimodal networks, each depicting the relationships between chemicals and genes. Each gene and its connections to its respective chemicals were color-coded based on the gene’s biological function according to KEGG pathways and the Reactome Knowledgebase (Kanehisa et al. 2008; Jassal et al. 2020). Nodes and edges are sized by weighted degree centrality. Larger nodes indicate more relevant data associated with a chemical or gene in the CTD-curated literature (Pinkhasova et al. 2021).
Caenorhabditis elegans culture
The C. elegans strain, Bristol N2, was procured from the Caenorhabditis Genetics Centre (University of Minnesota, MN, United States) and grown on OP50-seeded nematode growth medium (NGM) at 22 °C (Stiernagle 2006). Age-synchronization was achieved by the sodium hypochlorite method followed by overnight incubation of eggs in M9 buffer as described previously (Fabian and Johnson 1994).
Chemical exposure
Exposures of chlorpyrifos, permethrin, THC, and CBD were performed with ∼200 worms in 500 µl liquid media (M9 with cholesterol) for 48 h. In-well doses of 0, 7, 15, and 50 µM were used for chlorpyrifos; 0, 7.5, 15, 50, 100, and 200 µM for permethrin; 0, 7.5, 15, 50, and 100 µM for Δ9-THC; and 0, 7.5, 15, 50, 100, and 200 µM for CBD. The maximum concentrations indicated the limits for lethality (50 µM for chlorpyrifos) or solubility (200 µM for permethrin, 100 µM for Δ9-THC, and 200 µM for CBD). The worms were counted before treatment to be approximately 200 worms per well. Afterward, the 1,000× stocks of chlorpyrifos and permethrin in ethanol were added to the liquid media. The final in-well concentration of ethanol was 0.1% (v/v). Δ9-THC and CBD stocks were added to the liquid media at the beginning and re-dosed at 24-h time point. The final in-well concentration of ethanol was 0.2% (v/v). The re-dosing of Δ9-THC and CBD was necessary potentially due to the in-well degradation of cannabinoids (data not shown).
1-Nonanol repulsion behavioral assay
The repulsive C. elegans behavior to the odorant chemical 1-nonanol is dopamine-dependent (Kimura et al. 2010; Baidya et al. 2014; Smita et al. 2017; Sammi et al. 2018). It can be used as an indirect measure of dopamine levels as previously described (Sammi et al. 2023). Briefly, treated worms were washed 3 times with M9 buffer. Worms were placed on NGM plates. The poking lash dipped in 1-nonanol (Sigma Aldrich) was placed close to the head of the worms with ∼20 worms per treatment. The response of worms was video-recorded using a Leica S9i microscope. The time taken for the worms to show repulsive behavior was counted using a stopwatch. Any worm prodded accidentally was disregarded.
Statistical analysis
Data were analyzed using the Statistical Package for the Social Sciences (SPSS) version 27.0. First, the repulsion time was normalized with log base 2 transformation. For the single-chemical experiments, a 1-factor ANOVA was used to determine the dose-dependent effect each pesticide and cannabinoid had on the repulsion time of the worms. If the chemical exposure has a significant effect (P < 0.05), a Dunnett’s post hoc test will be conducted to compare each dose to control. For the co-exposure experiments, a 2-factor ANOVA was used to examine the dose-dependent effect of a pesticide with and without cannabinoids. The first factor of the analysis consisted of 4 doses with 0, 7.5, 15, and 50 µM for chlorpyrifos and 0, 25, 60, and 200 µM for permethrin. The second factor consisted of 3 groups—100 µM Δ9-THC, 100 µM CBD, and ethanol (i.e. no cannabinoid) as control. If the pesticide doses and cannabinoid groups have a significant interaction (P < 0.05), a pairwise comparison with Bonferroni multiple testing correction was conducted to compare each dose to control in each cannabinoid group.
Results
More jurisdictions are regulating organophosphate and pyrethroid pesticides in cannabis in the United States
In 2019, 32 states and Washington D.C. had legalized medical or recreational cannabis programs (Table 1). Pesticide residues were regulated in cannabis in 17 of those 33 jurisdictions. One jurisdiction specified pesticides for cannabis testing but provided no regulatory action levels. Another jurisdiction cited 40 C.F.R. 180 as regulatory action levels. All the remaining 15 jurisdictions provided regulatory action levels for at least one pyrethroid pesticide. In comparison, only 12 of the 15 jurisdictions provided regulatory action levels for at least one organophosphate pesticide. In 2023, 6 more jurisdictions had legalized cannabis programs. Pesticide regulations were found in 34 of the 39 legalized jurisdictions. Five jurisdictions specified no pesticide and 5 jurisdictions cited 40 C.F.R. 180 in their regulatory documents. Twenty-nine jurisdictions provided regulatory action levels for at least one pyrethroid pesticide, whereas 25 specified at least one organophosphate pesticide. The distribution of the numbers of pyrethroid and organophosphate pesticides specified by each jurisdiction is shown in Fig. 1. The medians of pyrethroid and organophosphate pesticides specified by each jurisdiction increased from zero pesticide in 2019 to 4.5 pyrethroid and 7 organophosphate pesticides in 2023, respectively. The sources for the pesticide regulation analysis were provided in Supplementary Material 1.
| Number of jurisdictions | ||
|---|---|---|
| 2019 | 2023 | |
| Legalized medical and recreational cannabis | 11 (Alaska, California, Colorado, Maine, Massachusetts, Michigan, Nevada, Oregon, Vermont, Washington, Washington D.C.) | 23 (Alaska, Arizona, California, Colorado, Connecticut, Delaware, Illinois, Maine, Maryland, Massachusetts, Michigan, Missouri, Montana, Nevada, New Jersey, New Mexico, New York, Oregon, Rhode Island, Vermont, Virginia, Washington, Washington D.C.) |
| Legalized medical cannabis only | 22 (Arizona, Arkansas, Connecticut, Delaware, Florida, Hawaii, Illinois, Louisiana, Maryland, Minnesota, Missouri, Montana, New Hampshire, New Jersey, New Mexico, New York, North Dakota, Ohio, Oklahoma, Pennsylvania, Rhode Island, West Virginia) | 16 (Alabama, Arkansas, Florida, Hawaii, Kentucky, Louisiana, Minnesota, Mississippi, New Hampshire, North Dakota, Ohio, Oklahoma, Pennsylvania, South Dakota, Utah, West Virginia) |
| Regulating pesticide residues in cannabis | 17 | 34 |
| Specifying action levels for pesticides regulationa | 15 | 29 |
| Specifying action levels for at least one organophosphate pesticidea | 12 | 25 |
| Specifying action levels for at least one pyrethroid pesticidea | 15 | 29 |
| Citing the subpart C of the U.S. Environmental Protection Agency’s regulations for tolerances and exemptions (40 C.F.R. 180) without specifying the jurisdiction’s own action levels for pesticide regulation | 1 | 5 |
Different jurisdictions specify different organophosphate and pyrethroid pesticides in their regulatory documents. The regulatory action levels of the pesticides most commonly specified by the jurisdictions were shown in Fig. 2. The pesticides were ranked by total numbers of the regulating jurisdictions in 2019 and 2023. The number of regulating jurisdictions increased for each pesticide over the 5-yr time span. The total range of regulatory action levels for both organophosphates and pyrethroids has remained consistent from 2019 to 2023, but the inner quartile range shrank in almost half of the cases (5 total), whereas it only grew in one instance (chlorpyrifos). This shows that over the years, state-level regulatory agencies have approached a consensus in regulatory action levels for organophosphate and pyrethroid pesticides.
Literature analysis returns connections of organophosphate pesticides, pyrethroid pesticides, and cannabinoids at the gene level
We identified 55 chemicals among 3 chemical classes in the CTD—including 14 cannabinoids, 30 organophosphate pesticides, and 11 pyrethroid pesticides—with 95 genes, 341 phenotypes, and the outcome of “Parkinson disease” and all descendent terms (MESH: D010300) in 563 computationally generated CGPD-tetramer constructs and 2,674 inferred CGPD-tetramers (3,237 in total; Table 2 and Supplementary Material 2). Nineteen organophosphate and 7 pyrethroid pesticides (63% and 64%, respectively) were regulated in cannabis by at least one US jurisdiction between 2019 and 2023. Dronabinol (389 tetramers), chlorpyrifos (1,019 tetramers), and decamethrin (269 tetramers) had the highest number of tetramers among their respective chemical classes (Fig. S1). The top gene targets with the highest number of tetramers are tumor necrosis factor (TNF; 190 tetramers), dopamine receptor D2 (DRD2; 189 tetramers), Parkinsonism-associated deglycase (PARK7; 177 tetramers), and superoxide dismutase 1 (SOD1; 149 tetramers).
| Tetramers | Chemicals | Regulated | Genes | ||
|---|---|---|---|---|---|
| Curated | Organophosphate pesticide | 427 | 3 | 3 | 54 |
| Pyrethroid pesticide | 136 | 2 | 2 | 27 | |
| Cannabinoid | 0 | 0 | N/A | 0 | |
| Inferred | Organophosphate pesticide | 1,172 | 30 | 19 | 76 |
| Pyrethroid pesticide | 770 | 11 | 7 | 68 | |
| Cannabinoid | 732 | 14 | N/A | 60 | |
| Total | 3,237 | 55 | 26 | 95 |
Figure 3A shows 351 chemical-gene connections that involved 2,331 tetramers, 30 organophosphate pesticides, 14 cannabinoids, and 88 genes. These connections were associated with 5 predicted mechanisms of action: Metabolism (19 genes), signal transduction (18 genes), genetic predisposition of Parkinson’s disease (13 genes), immune function (8 genes), and transmembrane transport (6 genes). Of those 88 genes, 48 are connected to both organophosphates and cannabinoids in Fig. 3A. These shared genes were associated with metabolism (12 genes), signal transduction (8 genes), genetic predisposition of Parkinson’s disease (9 genes), immune function (7 genes), and transmembrane transport (1 genes). Predicted mechanisms from this study can inform the selection of specific NAM technology to further examine how pesticides and cannabinoids may pose harms to individuals with Parkinson’s disease.
Figure 3B shows 303 chemical-gene connections that involved 1,638 tetramers related to 11 pyrethroid pesticides, 14 cannabinoids, and 85 genes. These connections were associated with metabolism (19 genes), signal transduction (19 genes), genetic predisposition of Parkinson’s disease (13 genes), immune function (8 genes), and transmembrane transport (9 genes). Of those 85 genes, 43 are connected to both OPs and cannabinoids in Fig. 3B. These shared genes were also associated with metabolism (9 genes), signal transduction (7 genes), genetic predisposition of Parkinson’s disease (9 genes), immune function (6 genes), and transmembrane transport (3 genes).
Exposure to chlorpyrifos and permethrin, but not Δ9-THC and CBD, results in dose-dependent effects on 1-nonanol repulsive behaviors
The main effects of chlorpyrifos, permethrin, Δ9-THC, and CBD on 1-nonanol repulsive behaviors in C. elegans were examined using 4 separate 1-way ANOVAs (Fig. 4A, n = 93 to 103 for chlorpyrifos; 4B, 36 to 45 for permethrin; 4C, 39 to 46 for Δ9-THC; 4D, 37 to 44 for CBD). For chlorpyrifos, there was a statistically significant difference between the dose groups (P < 0.001). A Dunnett post hoc test showed that the 15 and 50 µM doses were significantly different from the control (P < 0.001 for both). There was no significant difference between the 7.5 µM dose and the control (P > 0.05). For permethrin, there was a statistically significant difference between the dose groups (P = 0.002). A Dunnett post hoc test showed that the 200-µM dose is significantly different from the control (P = 0.006). There was no statistically significant difference between the lower doses (7.5 to 100 µM) of permethrin and the control (P > 0.05). For Δ9-THC and CBD, there was not a statistically significant difference between the dose groups (P > 0.05).
Dose-dependent effects of chlorpyrifos are different in the presence of Δ9-THC and CBD
The interaction of chlorpyrifos and cannabinoid (none, Δ9-THC, or CBD) co-exposure on 1-nonanol repulsive behaviors in C. elegans was examined using 2 separate 2-way ANOVAs (Fig. 5; n = 93 to 103 for chlorpyrifos). Chlorpyrifos doses and cannabinoid groups showed a statistically significant interaction (P < 0.001). Bonferroni’s post hoc tests for pairwise comparisons showed significant differences between the control and 50 µM in all 3 cannabinoid groups. The control and 15 µM were significantly different with CBD and without cannabinoid, but not with Δ9-THC. Additionally, the control and 7.5 µM were significantly different with CBD, but not with Δ9-THC and without cannabinoid. There was also statistically significant difference between the controls in the Δ9-THC and no-cannabinoid groups (P < 0.01). However, the 1.3- to 2-s difference in response time was below the limit of detection in this assay (data not shown). Taken together, the result indicated different dose-dependent effects of chlorpyrifos in the presence of Δ9-THC and CBD, which showed no significant effects by themselves (Fig. 4C and D). The interaction between chlorpyrifos doses and the endocannabinoid 2-arachidonoylglycerol was also demonstrated using fluorescent microscopy (Fig. S2). The effects of permethrin were close to the limit of detection in this assay. Therefore, it was not included in the interaction study.
Discussion
In this study, we found growing consensus in state-level regulations of organophosphate and pyrethroid pesticides in cannabis. Over the past 5 yr, the number of jurisdictions regulating organophosphate and pyrethroid pesticides increased in the United States. This could result in an increase in contaminated cannabis getting removed from the market due to these 2 pesticide classes (Jameson et al. 2022). Additionally, the regulatory action levels of organophosphate and pyrethroid pesticides were harmonizing in different states. As more states were legalizing medical cannabis (Griffith et al. 2024), there was more collaboration and discussion between state agencies and industry practitioners through government and professional associations, such as the Cannabis Regulators Association, the Association of Public Health Laboratories, the U.S. Pharmacopeia, the ASTM, and the AOAC International. Cannabis production and analysis companies that operate in multiple states (collectively known as multi-state operators) also share expertise, personnel, and concerns across state lines. The activities of these stakeholders may account for the observed harmonization of cannabis pesticide regulations in different states.
Although public and industry sectors are harmonizing cannabis pesticide regulations based on consensus and production experience, more research remains needed to support human health risk assessment and develop health-protective action levels. As a proof of concept, this study demonstrated how to use NAMs to generate predictions and strategize for mechanistic testing of cannabinoid-pesticide mixture. The top 2 predicted mechanisms of action for pesticide-cannabinoid mixtures are metabolism and signal transduction. This is consistent with other reports on potential drug interaction and endocannabinoid signaling disruption of cannabis use (Stout and Cimino 2014; Leung et al. 2019). Further NAM studies can explore the plausible metabolic interactions and safety of pesticide-cannabinoid mixtures based on physiologically based pharmacokinetic modeling and qualitative structure–activity relationship (Arvidson 2012; Rowland Yeo et al. 2024). There are also NAM technologies focusing on specific mechanisms of action—such as signal transduction (Bundy et al. 2024), immune functions (Snapkow et al. 2024), and transmembrane transport (Dolghih et al. 2011)—to generate new information on plausible interactions of pesticides and cannabinoids.
The CTD analysis identified the potential contribution of genetic predisposition in pesticide-cannabinoid mixture toxicity. Genetic backgrounds contribute to many neurological conditions where medical cannabis is used as a treatment option (Pinkhasova et al. 2021). The current study predicts 13 genes related to Parkinson’s disease with connections to pesticides or cannabinoids. Additionally, 9 genes are associated with transmembrane transport, which is a key biological function linked to Parkinson’s disease (Abeliovich and Gitler 2016). Many of these genes are also targets of pesticide exposure (Chedik et al. 2018). Genetic predisposition may worsen Parkinson’s disease when combined with pesticide exposure (Brown et al. 2024). Although this study did not demonstrate how each pesticide modulates each gene target individually, it provided new hypotheses for further examination of the role of genetic variability in pesticide-cannabinoid mixture toxicity.
By combining the knowledge from different model systems, the CTD analysis also generated a new mechanistic hypothesis where pesticides and cannabinoids lead to oxidative stress and subsequent mitochondrial dysfunction in dopaminergic neurons. The top 4 genes with the most connections in the CTD are TNF, DRD2, PARK7, and SOD1, 3 of which (DRD2, PARK7, and SOD1) have functional homologs in C. elegans and have been previously studied for mitochondrial dopaminergic toxicity (Nass and Blakely 2003; Wang et al. 2009; Cooper and Van Raamsdonk 2018). This prediction is consistent with earlier studies on Δ9-THC, CBD, chlorpyrifos, and permethrin (Singh et al. 2015, 2018; Wang et al. 2016; Chan and Duncan 2021) and the current study in C. elegans, where the organophosphate pesticide chlorpyrifos and cannabinoids interacted in depleting dopamine in dopaminergic neurons. Although this interaction is yet to be confirmed in a more relevant model to humans, the use of transgenic C. elegans with knock-in human genes can provide additional evidence. Furthermore, chlorpyrifos and many organophosphate pesticides can all affect mitochondrial function in dopaminergic neurons (Leung and Meyer 2019; Sammi et al. 2023). Further studies are needed to evaluate mitochondrial toxicity upon exposure to multiple organophosphate pesticides in contaminated cannabis products.
This study demonstrated the advantage of conducting these C. elegans assays where large numbers of individuals can be tested in a quick and relatively cheap fashion. This is critical when testing mixtures of potential contaminants where the large number of possible permutations of chemical combinations are simply not feasible with larger and more expensive test organisms. Although the C. elegans studies did not show conclusive evidence of synergistic effects between chlorpyrifos and cannabinoids as expected, we demonstrated a change in dose–response relationship of a pesticide in a botanical commodity that is neuroactive. This highlights a key difference in the hazard assessment of a pesticide in a conventional agricultural commodity versus a botanical commodity like cannabis. The pharmacological properties of the botanical commodity itself can add to the hazard of the contaminant being assessed. Further toxicology studies need to consider the potential interaction between pesticides and cannabinoids (as well as other neuroactive components in cannabis) in determining the point of departure for human health risk assessment.
In conclusion, this study provides an important proof of concept for the use of NAMs to inform further testing and pesticide regulations in cannabis. The CTD analysis leverages the current literature to predict specific mechanisms of action in which pesticides and cannabinoids may interact, thereby informing NAM technology selection for fit-for-purpose testing (i.e. using C. elegans models for pesticide exposure in cannabis for individuals with Parkinson’s disease in this example). This approach also informs cannabis pesticide risk assessment by forecasting plausible gene–environment interaction in such exposure scenarios. Lastly, this study characterizes the mechanisms of action of cannabinoids combined with specific pesticide classes. This chemical class-based approach to risk assessment can provide a solution to evaluate a large combination of cannabis-pesticide mixtures and refine cannabis pesticide regulations at scale.
Supplementary Material
Acknowledgments
We would like to thank Steven Baker, Mary Graham, and the Arizona Department of Health Services for providing the 2019 data of state-level pesticide regulations in the United States. Δ9-tetrahydrocannabinol was provided by the Drug Supply Program of the National Institute on Drug Abuse.
Contributor Information
Albert B Rivera, School of Mathematical and Natural Sciences, Arizona State University, Glendale, AZ 85306, United States; ASU-Banner Neurodegenerative Disease Research Center, Arizona State University, Tempe, AZ 85281, United States.
Ariell B Stephens, School of Mathematical and Natural Sciences, Arizona State University, Glendale, AZ 85306, United States.
Kendra D Conrow, School of Mathematical and Natural Sciences, Arizona State University, Glendale, AZ 85306, United States.
Symone T Griffith, ASU-Banner Neurodegenerative Disease Research Center, Arizona State University, Tempe, AZ 85281, United States; College of Health Solutions, Arizona State University, Phoenix, AZ 85004, United States.
Laura E Jameson, School of Mathematical and Natural Sciences, Arizona State University, Glendale, AZ 85306, United States.
Thomas M Cahill, School of Mathematical and Natural Sciences, Arizona State University, Glendale, AZ 85306, United States.
Shreesh R Sammi, Department of Translational Neuroscience, Michigan State University, Grand Rapids, MI 49503, United States.
Mathew R Swinburne, Francis King Carey School of Laws, University of Maryland, Baltimore, MD 21201, United States.
Jason R Cannon, School of Health Sciences, Purdue University, West Lafayette, IN 47907, United States; Purdue Institute for Integrative Neuroscience, Purdue University, West Lafayette, IN 47907, United States.
Maxwell C K Leung, School of Mathematical and Natural Sciences, Arizona State University, Glendale, AZ 85306, United States; ASU-Banner Neurodegenerative Disease Research Center, Arizona State University, Tempe, AZ 85281, United States; College of Health Solutions, Arizona State University, Phoenix, AZ 85004, United States.
Supplementary material
Supplementary material is available at Toxicological Sciences online.
Funding
None declared.
Conflicts of interest. None declared.
References
Untitled section
References
- Abeliovich A, Gitler AD. 2016. Defects in trafficking bridge Parkinson’s disease pathology and genetics. Nature. 539:207–216. 10.1038/nature20414
- Arizona Department of Health Services (AZDHS), Compiled State Pesticides/Fungicides/Herbicides/Growth Regulators Review. 2019. Medical Marijuana Testing Advisory Council. Phoenix (AZ). [accessed 2024 August 9]. https://www.azdhs.gov/licensing/medical-marijuana/index.php#testing-advisory-council
- Arvidson KB. 2012. QSAR and Read-Across Approaches using in silico Tools in Food Ingredient/Contaminant Safety Assessments at the U.S. Food & Drug Administration. Cosmetic Ingredient Review. Silver Spring (MD): U.S. Food & Drug Administration. [accessed 2024 August 9]. https://www.cir-safety.org/supplementaldoc/qsar-and-read-across-approaches-using-silico-tools-food-ingredient/contaminant-safet
- Baidya M, Genovez M, Torres M, Chao MY. 2014. Dopamine modulation of avoidance behavior in Caenorhabditis elegans requires the NMDA receptor NMR-1. PLoS One. 9:e102958. 10.1371/journal.pone.0102958
- Bastian M, Heymann S, Jacomy M. 2009. Gephi: an open source software for exploring and manipulating networks. ICWSM. 3:361–362. 10.1609/icwsm.v3i1.13937
- Brown EG, Goldman SM, Coffey CS, Siderowf A, Simuni T, Meng C, Brumm MC, Caspell-Garcia C, Marek K, Tanner CM; Parkinson’s Progression Markers Initiative. 2024. Occupational pesticide exposure in Parkinson’s disease related to GBA and LRRK2 variants. J Parkinsons Dis. 14:737–746. 10.3233/JPD-240015
- Bundy JL, Everett LJ, Rogers JD, Nyffeler J, Byrd G, Culbreth M, Haggard DE, Word LJ, Chambers BA, Davidson-Fritz S, et al. 2024. High-throughput transcriptomics screen of ToxCast chemicals in U-2 OS cells. Toxicol Appl Pharmacol. 491:117073. 10.1016/j.taap.2024.117073
- Chan JZ, Duncan RE. 2021. Regulatory effects of cannabidiol on mitochondrial functions: a review. Cells. 10:1251. 10.3390/cells10051251
- Chedik L, Bruyere A, Bacle A, Potin S, Le Vée M, Fardel O. 2018. Interactions of pesticides with membrane drug transporters: implications for toxicokinetics and toxicity. Expert Opin Drug Metab Toxicol. 14:739–752. 10.1080/17425255.2018.1487398.
- Cooper JF, Van Raamsdonk JM. 2018. Modeling Parkinson’s disease in C. elegans. J Parkinsons Dis. 8:17–32. 10.3233/JPD-171258
- Curated [chemical–gene interactions|chemical–disease|gene–disease] data were retrieved from the Comparative Toxicogenomics Database (CTD). 2023. MDI Biological Laboratory, Salisbury Cove, Maine, and NC State University, Raleigh, North Carolina. World Wide Web (https://ctdbase.org/) [2023 Nov 30, revision 17204].
- Davis AP, Grondin CJ, Johnson RJ, Sciaky D, Wiegers J, Wiegers TC, Mattingly CJ. 2021. Comparative toxicogenomics database (CTD): update 2021. Nucleic Acids Res. 49:D1138–D1143. 10.1093/nar/gkaa891
- Davis AP, Wiegers TC, Wiegers J, Wyatt B, Johnson RJ, Sciaky D, Barkalow F, Strong M, Planchart A, Mattingly CJ. 2023. CTD tetramers: a new online tool that computationally links curated chemicals, genes, phenotypes, and diseases to inform molecular mechanisms for environmental health. Toxicol Sci. 195:155–168. 10.1093/toxsci/kfad069
- Dolghih E, Bryant C, Renslo AR, Jacobson MP. 2011. Predicting binding to p-glycoprotein by flexible receptor docking. PLoS Comput Biol. 7:e1002083. 10.1371/journal.pcbi.1002083
- Dryburgh LM, Bolan NS, Grof CPL, Galettis P, Schneider J, Lucas CJ, Martin JH. 2018. Cannabis contaminants: sources, distribution, human toxicity and pharmacologic effects. Br J Clin Pharmacol. 84:2468–2476. 10.1111/bcp.13695.
- Fabian TJ, Johnson TE. 1994. Production of age-synchronous mass cultures of Caenorhabditis elegans. J Gerontol. 49:B145–B156. 10.1093/geronj/49.4.b145
- Feeney MP, Bega D, Kluger BM, Stoessl AJ, Evers CM, De Leon R, Beck JC. 2021. Weeding through the haze: a survey on cannabis use among people living with Parkinson’s disease in the US. NPJ Parkinsons Dis. 7:21. 10.1038/s41531-021-00165-y
- Figura M, Koziorowski D, Sławek J. 2022. Cannabis in Parkinson’s disease–the patient’s perspective versus clinical trials: a systematic literature review. Neurol Neurochir Pol. 56:21–27. 10.5603/PJNNS.a2022.0004
- Furlong M, Tanner CM, Goldman SM, Bhudhikanok GS, Blair A, Chade A, Comyns K, Hoppin JA, Kasten M, Korell M, et al. 2015. Protective glove use and hygiene habits modify the associations of specific pesticides with Parkinson’s disease. Environ Int. 75:144–150. 10.1016/j.envint.2014.11.002.
- Griffith ST, Conrow KD, Go M, McEntee ML, Daniulaityte R, Nadesan MH, Swinburne MR, Shill HA, Leung MCK. 2024. Cannabis use in Parkinson’s disease: patient access to medical cannabis and physician perspective on product safety. Neurotoxicology. 103:198–205. 10.1016/j.neuro.2024.05.008
- Jameson LE, Conrow KD, Pinkhasova DV, Boulanger HL, Ha H, Jourabchian N, Johnson SA, Simeone MP, Afia IA, Cahill TM, et al. 2022. Comparison of state-level regulations for cannabis contaminants and implications for public health. Environ Health Perspect. 130:97001. 10.1289/EHP11206
- Jassal B, Matthews L, Viteri G, Gong C, Lorente P, Fabregat A, Sidiropoulos K, Cook J, Gillespie M, Haw R, et al. 2020. The Reactome Pathway Knowledgebase. Nucleic Acids Res. 48:D498–D503. 10.1093/nar/gkz1031
- Kanehisa M, Araki M, Goto S, Hattori M, Hirakawa M, Itoh M, Katayama T, Kawashima S, Okuda S, Tokimatsu T, et al. 2008. KEGG for linking genomes to life and the environment. Nucleic Acids Res. 36:D480–D484. 10.1093/nar/gkm882
- Kavlock RJ, Bahadori T, Barton-Maclaren TS, Gwinn MR, Rasenberg M, Thomas RS. 2018. Accelerating the pace of chemical risk assessment. Chem Res Toxicol. 31:287–290. 10.1021/acs.chemrestox.7b00339
- Kimura KD, Fujita K, Katsura I. 2010. Enhancement of odor avoidance regulated by dopamine signaling in Caenorhabditis elegans. J Neurosci. 30:16365–16375. 10.1523/JNEUROSCI.6023-09.2010
- Kindred JH, Li K, Ketelhut NB, Proessl F, Fling BW, Honce JM, Shaffer WR, Rudroff T. 2017. Cannabis use in people with Parkinson’s disease and multiple sclerosis: a web-based investigation. Complement Ther Med. 33:99–104. 10.1016/j.ctim.2017.07.002
- Kortenkamp A, Faust M. 2018. Regulate to reduce chemical mixture risk. Science. 361:224–226. 10.1126/science.aat9219
- Leung MCK, Meyer JN. 2019. Mitochondria as a target of organophosphate and carbamate pesticides: revisiting common mechanisms of action with new approach methodologies. Reprod Toxicol. 89:83–92. 10.1016/j.reprotox.2019.07.007
- Leung MCK, Silva MH, Palumbo AJ, Lohstroh PN, Koshlukova SE, DuTeaux SB. 2019. Adverse outcome pathway of developmental neurotoxicity resulting from prenatal exposures to cannabis contaminated with organophosphate pesticide residues. Reprod Toxicol. 85:12–18. 10.1016/j.reprotox.2019.01.004
- Leung MCK, Williams PL, Benedetto A, Au C, Helmcke KJ, Aschner M, Meyer JN. 2008. Caenorhabditis elegans: an emerging model in biomedical and environmental toxicology. Toxicol Sci. 106:5–28. 10.1093/toxsci/kfn121.
- Li S, Ritz B, Gong Y, Cockburn M, Folle AD, Del Rosario I, Yu Y, Zhang K, Castro E, Keener AM, et al. 2023. Proximity to residential and workplace pesticides application and the risk of progression of Parkinson’s diseases in Central California. Sci Total Environ. 864:160851. 10.1016/j.scitotenv.2022.160851
- Lu S, Liu SS, Huang P, Wang ZJ, Wang Y. 2021. Study on the combined toxicities and quantitative characterization of toxicity sensitivities of three flavor chemicals and their mixtures to Caenorhabditis elegans. ACS Omega. 6:35745–35756. 10.1021/acsomega.1c05688
- Narayan S, Liew Z, Paul K, Lee PC, Sinsheimer JS, Bronstein JM, Ritz B. 2013. Household organophosphorus pesticide use and Parkinson’s disease. Int J Epidemiol. 42:1476–1485. 10.1093/ije/dyt170
- Nass R, Blakely RD. 2003. The Caenorhabditis elegans dopaminergic system: opportunities for insights into dopamine transport and neurodegeneration. Annu Rev Pharmacol Toxicol. 43:521–544. 10.1146/annurev.pharmtox.43.100901.135934
- National Conference of State Legislatures (NCSL). 2024. State Medical Cannabis Laws. https://www.ncsl.org/health/state-medical-cannabis-laws
- Pinkhasova DV, Jameson LE, Conrow KD, Simeone MP, Davis AP, Wiegers TC, Mattingly CJ, Leung MCK. 2021. Regulatory status of pesticide residues in cannabis: implications to medical use in neurological diseases. Curr Res Toxicol. 2:140–148. 10.1016/j.crtox.2021.02.007
- Pruyn SA, Wang Q, Wu CG, Taylor CL. 2022. Quality standards in state programs permitting cannabis for medical uses. Cannabis Cannabinoid Res. 7:728–735. 10.1089/can.2021.0164
- Rager JE, Lichtveld K, Ebersviller S, Smeester L, Jaspers I, Sexton KG, Fry RC. 2011. A toxicogenomic comparison of primary and photochemically altered air pollutant mixtures. Environ Health Perspect. 119:1583–1589. 10.1289/ehp.1003323.
- Rowland Yeo K, Gil Berglund E, Chen Y. 2024. Dose optimization informed by PBPK modeling: state-of-the art and future. Clin Pharmacol Ther. 116:563–576. 10.1002/cpt.3289
- Sammi SR, Agim ZS, Cannon JR. 2018. From the cover: Harmane-induced selective dopaminergic neurotoxicity in Caenorhabditis elegans. Toxicol Sci. 161:335–348. 10.1093/toxsci/kfx223
- Sammi SR, Syeda T, Conrow KD, Leung MCK, Cannon JR. 2023. Complementary biological and computational approaches identify distinct mechanisms of chlorpyrifos versus chlorpyrifos-oxon-induced dopaminergic neurotoxicity. Toxicol Sci. 191:163–178. 10.1093/toxsci/kfac114
- Schauer GL, Johnson JK, Rak DJ, Dodson L, Steinfeld N, Sheehy TJ, Nakata M, Collins SP. 2023. A research agenda to inform cannabis regulation: how science can shape policy. Clin Ther. 45:506–514. 10.1016/j.clinthera.2023.03.010
- Serafini MM, Sepehri S, Midali M, Stinckens M, Biesiekierska M, Wolniakowska A, Gatzios A, Rundén-Pran E, Reszka E, Marinovich M, et al. 2024. Recent advances and current challenges of new approach methodologies in developmental and adult neurotoxicity testing. Arch Toxicol. 98:1271–1295. 10.1007/s00204-024-03703-8
- Singh N, Hroudová J, Fišar Z. 2015. Cannabinoid-induced changes in the activity of electron transport chain complexes of brain mitochondria. J Mol Neurosci. 56:926–931. 10.1007/s12031-015-0545-2
- Singh N, Lawana V, Luo J, Phong P, Abdalla A, Palanisamy B, Rokad D, Sarkar S, Jin H, Anantharam V, et al. 2018. Organophosphate pesticide chlorpyrifos impairs STAT1 signaling to induce dopaminergic neurotoxicity: implications for mitochondria mediated oxidative stress signaling events. Neurobiol Dis. 117:82–113. 10.1016/j.nbd.2018.05.019
- Smita SS, Sammi SR, Laxman TS, Bhatta RS, Pandey R. 2017. Shatavarin IV elicits lifespan extension and alleviates parkinsonism in Caenorhabditis elegans. Free Radic Res. 51:954–969. 10.1080/10715762.2017.1395419
- Snapkow I, Smith NM, Arnesdotter E, Beekmann K, Blanc EB, Braeuning A, Corsini E, Sollner Dolenc M, Duivenvoorde LPM, Sundstøl Eriksen G, et al. 2024. New approach methodologies to enhance human health risk assessment of immunotoxic properties of chemicals—a PARC (partnership for the assessment of risk from chemicals) project. Front Toxicol. 6:1339104. 10.3389/ftox.2024.1339104
- St John P, Halperin A. 2024. The Dirty Secret of California’s Legal Weed. Los Angeles Times. [accessed 2024 August 9]. https://www.latimes.com/california/story/2024-06-14/the-dirty-secret-of-californias-legal-weed
- Stiernagle T. 2006. Maintenance of C. elegans. WormBook. The C. elegans Research Community, WormBook. 10.1895/wormbook.1.101.1. [accessed 2024 August 9]. http://www.wormbook.org
- Stout SM, Cimino NM. 2014. Exogenous cannabinoids as substrates, inhibitors, and inducers of human drug metabolizing enzymes: a systematic review. Drug Metab Rev. 46:86–95. 10.3109/03602532.2013.849268
- Tang B, Tong P, Xue KS, Williams PL, Wang JS, Tang L. 2019. High-throughput assessment of toxic effects of metal mixtures of cadmium(Cd), lead(Pb), and manganese(Mn) in nematode Caenorhabditis elegans. Chemosphere. 234:232–241. 10.1016/j.chemosphere.2019.05.271.
- U.S. Environmental Protection Agency (U.S. EPA). 2024. Tolerances and exemptions for pesticide chemical residues in food. Washington, D.C.: 40 Code of Federal Regulations Part 180. [accessed 2024 August 9]. https://www.ecfr.gov/current/title-40/part-180
- Wang A, Cockburn M, Ly TT, Bronstein JM, Ritz B. 2014. The association between ambient exposure to organophosphates and Parkinson’s disease risk. Occup Environ Med. 71:275–281. 10.1136/oemed-2013-101394
- Wang J, Farr GW, Hall DH, Li F, Furtak K, Dreier L, Horwich AL. 2009. An ALS-linked mutant SOD1 produces a locomotor defect associated with aggregation and synaptic dysfunction when expressed in neurons of Caenorhabditis elegans. PLoS Genet. 5:e1000350. 10.1371/journal.pgen.1000350
- Wang X, Martínez M-A, Dai M, Chen D, Ares I, Romero A, Castellano V, Martínez M, Rodríguez JL, Martínez-Larrañaga M-R, et al. 2016. Permethrin-induced oxidative stress and toxicity and metabolism. A review. Environ Res. 149:86–104. 10.1016/j.envres.2016.05.003
- Wang Y, Liu SS, Huang P, Wang ZJ, Xu YQ. 2021. Assessing the combined toxicity of carbamate mixtures as well as organophosphorus mixtures to Caenorhabditis elegans using the locomotion behaviors as endpoints. Sci Total Environ. 760:143378. 10.1016/j.scitotenv.2020.143378
- Wittkowski P, Marx-Stoelting P, Violet N, Fetz V, Schwarz F, Oelgeschläger M, Schönfelder G, Vogl S. 2019. Caenorhabditis elegans as a promising alternative model for environmental chemical mixture effect assessment-a comparative study. Environ Sci Technol. 53:12725–12733. 10.1021/acs.est.9b03266.