Tween 80 (polysorbate 80) was obtained from Sigma-Aldrich (St. Louis, MO, United States), CAS number 9005-65-6. Sesame oil was obtained from Sigma-Aldrich, CAS number 8008-74-0, stored at 4°C and used within 3 months of opening the bottle. Purified cannabidiol (CBD) was obtained from Cayman Chemical (Ann Arbor, MI, United States), CAS number 13956-29-1, ≥98% purity. Isotopically labeled internal standard stock solution of CBD-D₃ (SKU: C-084, 0.1 mg/mL in methanol, Cerilliant^®) was purchased from Sigma-Aldrich. Acetonitrile (CHROMA-SOLV ™, LC-MS grade) was purchased from Fisher Scientific (Hampton, NH, United States). Ultrapure deionized water (resistivity 18.2 MΩ-cm) was produced using a Milli-Q system from Millipore Sigma (Burlington, MA, United States).

An ethanolic hemp extract was provided by the National Center for Natural Products Research at the University of Mississippi, University, MS, USA. The concentrations of cannabinoids in the extract were reported to be: tetrahydrocannabinol at 1.6 mg/mL, CBD at 52.6 mg/mL, cannabichromene at 1.5 mg/mL, tetrahydrocannabivarin at 0.1 mg/mL, cannabigerol at 1.3 mg/mL, and cannabinol at 0.2 mg/mL (Zhao et al., 2023). These concentrations were verified to be ≤7.1% difference prior to use. Ethanol was removed from the hemp extract using a CentriVap^® Centrifugal Concentrator from Labconco^® Corporation (Kansas City, MO). Two cycles of centrifugation with no heat were used to evaporate ethanol from the samples. Between each cycle, the samples were placed on ice for 10 min to cool. After centrifugation and evaporation, samples were reconstituted in sesame oil, and CBD concentration was measured by ultra-high performance liquid chromatography-electrospray tandem mass spectrometry (UHPLC-ESI MS/MS, Agilent Technologies, Santa Clara, CA) prior to the experiments. Verified concentrations were then used to calibrate hemp extract exposures by CBD concentration. Initial hemp extract resuspensions in sesame oil allowed for the maximum exposures of 960 μg/mL CBD used for the experiments described in Sections 3.2–3.4. Later preparations achieved slightly more concentrated hemp extract in sesame oil, allowing for the 1,000 μg/mL CBD exposures used for the experiments in Sections 3.5 and 3.6.

C. elegans strains CL2166 (gst-4p::GFP, oxidative stress response), PD4251 (myo-3p::GFP, progeny-to-adult ratio assay), and wild-type (N2, all other assays) were supplied by the Caenorhabditis Genetics Center (CGC), which is supported by the US National Institutes of Health—Office of Research Infrastructure Programs (P40 OD010440). All strains were initially grown on agar plates with a continuous ample E. coli food supply prior to hypochlorite treatment and transfer to C. elegans habitation medium (CeHM) —a mixture of 80% chemically defined nutrient medium and 20% non-fat cows’ milk (Clegg et al., 2002; Sprando et al., 2009). CeHM was prepared as per Hunt et al. (2023). Prior to use in testing, frozen strains were maintained in CeHM in vented flasks on orbital shakers inside hot/cold incubators set to 20°C for a minimum of 3 weeks with continuous ample food supply. To avoid genetic drift, strain aliquots were utilized for a maximum of 6 months prior to acclimating a fresh aliquot that had been frozen within a few weeks of receipt from the repository. Crowded culture flasks or those with dauers were discarded. C. elegans males spontaneously arise only rarely and do not seem to propagate in liquid culture on shakers; therefore, exposed C. elegans were presumed to be hermaphrodites. Culture maintenance and exposures were performed in incubators set to 20°C unless otherwise noted.

Stable emulsions can be prepared by adding an emulsifier and a surfactant to a mixture of oil-in-water (Zhang et al., 2020). For the emulsions in this study, the proteins in cow’s milk acted as emulsifiers, and the Tween 80 (Polysorbate 80) as a food-grade surfactant. An example emulsion preparation worksheet is provided as Supplementary Table 2.

Step 1, mixture preparation: 1 mL of either a) sesame oil, b) purified CBD in sesame oil, or c) hemp extract of predetermined CBD concentration in sesame oil was mixed with 19 mL of a mixture of 1% Tween 80 in non-fat cows’ milk (1%Tw80).

Emulsion particle size and zeta potential were measured with a ZetaPALS from Brookhaven Instruments (Holtsville, New York) using 92-nm latex particles and BI-ZR5 Zeta potential validation material (ζ = −44 ± 8 mV) from the same company as internal standards. The emulsions were utilized in such a way that experiments were completed within 8 days of emulsion preparation. No changes, such as an oily surface or precipitation in the media, were noted by eye or microscopy in exposed wells during the experiments or in the 10× dosing emulsions, which were maintained for up to 2 months at 4°C to assess stability. These observations are in contrast with initial emulsion preparations using lower psi or fewer passes through the microfluidizer, which resulted in a visible oily sheen at the top of the experimental wells within a day or two.

The transgenic C. elegans strain CL2166 (gst-4p::GFP) was used as a biomarker for the conserved Nrf2-mediated oxidative stress response. Synchronized cohorts were exposed for 24 h, from the onset of the young adult stage, as determined by at least 50% of the cohort having at least one visible primary oocyte but no internal fertilized eggs. This occurred at approximately 65 h post L1 feeding in both wild-type and CL2166 worms grown at 20°C in CeHM. The effects on C. elegans transgene expression, size, and optical density were evaluated with a COPAS^® (Union Biometrica, Holliston, MA). Because the progeny can be easily gated separately from the original exposed cohort, no contraceptive was needed. Morphometric time-of-flight (TOF) measurements can be used as phenotypic comparisons to controls to define population health. Populations with mean TOF values that differed from matched controls by > 20% were excluded from analyses. The data shown are the mean green fluorescence values relative to the control and include no effect levels where available up to the highest concentration obtainable that maintained TOF within 80% of the control. Significance was determined from a minimum of four independent experiments per exposure group for Student’s T-test p-values *<0.05, # <0.005. Mean changes considered biologically meaningful were ≥10% for fluorescence and ≥5% for TOF.

See Hunt et al. (2023) for detailed methods for this assay. Briefly, synchronized N2 wild-type C. elegans were exposed to the contraceptive 5-fluoro-2′-deoxyuridine (FUdR) when at least 50% of the cohort had reached the mid-L4 stage. Adults were exposed the following day to purified CBD or CBD-rich hemp extract. Spontaneous locomotor activity levels were recorded for 18 h with a WMicroTracker™ (Phylumtech, Santa Fe, Argentina), an activity tracking instrument placed in a hot/cold incubator set to 20°C. Data were analyzed using the web tool https://waat.galaxytrakr.org. The data presented comprise five independently generated cohorts assessed with three or four independently produced emulsion preparations.

See Camacho et al. (2023) for detailed methods for this assay. Briefly, C. elegans strain PD4251 grows with wild-type parameters but glows a bright green from just before hatching and throughout adulthood, facilitating the gating of small progeny from the background with a COPAS. Stock cultures and exposed cohorts were maintained at 19°C so that all experimental steps could take place between 8 am and 8 pm Age-synchronized cohorts were continuously exposed via their nutrient media from the onset of reproductive maturity (the young adult stage), as determined by microscopy, when at least half of the population had a visible primary oocyte but no fertilized eggs. An example worksheet for this assay is provided as Supplementary Table 1. Replicate plates were evaluated by COPAS at 2, 3, and 4 days post-dosing. Significance was determined from a minimum of three independent experiments per exposure group for Student’s T-test p-values <0.05. TOF changes of <5% were within the variance between controls for this instrument and therefore were not considered biologically meaningful.

The wDAT uses wild-type N2 C. elegans to measure the timing of developmental milestone acquisition and stage-specific spontaneous locomotor activity. Age-synchronized cohorts were exposed from the onset of feeding after hatching and were tracked for 3 days through the four developmental stages in 12-well plates placed in wMicroTrackers. In some wDAT experiments, cohorts were exposed continuously for 3 days. In others, as indicated, the reversibility of the developmental effect was evaluated, with C. elegans exposed for 24.5 h +/-0.5 h from the onset of the initial L1 feeding. These are referred to in the text as “early-only” exposures. To avoid starvation during the washing steps, worms were washed twice, two wells at a time, followed by resuspension in CeHM within 10 min of commencing the washing steps. An example worksheet for this assay is provided as Supplementary Table 1.

For this study, L1 stage-specific locomotor activity peak times and heights were evaluated differently than previously described. When peak shapes are relatively uniform across exposures, as was the case for the L2, L3, and L4 values presented here, a mean of 3.5 h of wMT data is usually sufficient to smooth activity curves and determine peaks (Camacho et al., 2022). However, the L1 activity curves for higher concentrations of CBD and hemp extract were very different in shape from the control curves. Therefore, in order to accurately reflect the time and activity at the mid-L1 stage, the peak time for all wDAT exposures was reported as half of the time to the L1–L2 lethargus (determined as the lowest wMT value between the L1 and L2 stages), and the L1 peak height was measured as the mean of 7.5 h of wMT data centered evenly around this midpoint.

At the higher concentrations of CBD, there was some loss of synchronous developmental timing, which resulted in less clear drops in wDAT activity between developmental stages in some experiments, especially in the L4-adult lethargus. Data points where some peak times had to be estimated from the overall shape of the activity curve, rather than from a clear post-peak drop in activity as seen with lethargus in synchronized populations, are indicated with a “!” in the figures. Additionally, some experiments did not meet the acceptance criteria, either for the shape of the control activity curves and/or for the observed contamination. This may be due to the microfluidizer being too large to fit into a biological safety cabinet and the 3-day duration of the wDAT providing time for contaminants to grow. The number of experiments on which the results are based is indicated in the caption for each figure.

N2 C. elegans were exposed for 24 h at 20°C in 12-well plates with 900 µL CeHM plus 100 µL of 10× dosing emulsion per well, using either approximately 900 adults or 2,700 juveniles per well. L1 worms were counted right after the initial feeding, using a minimum of ten 5 µL samples; however, final worm numbers are estimates by necessity (Scanlan et al., 2018). Adults were exposed when at least 50% of the population had a visible internal oocyte, indicating that they had reached the young adult stage. Juveniles were exposed right after L1 feeding and counting. Adult plates were exposed on shakers to mimic the conditions used in the transgene and progeny ratio assay experiments. Juvenile plates were exposed without shaking, consistent with no shaking in the wDAT experiments. After 24 h, worm pellets were washed once each in 14 mL 0.1% Tween 80 in M9, 0.1% Tween 80 in water, and finally water. A total of 900µL of acetonitrile was added to the washed worm pellets in ∼100 µL of remaining water supernatant. To disrupt the cuticles, 2 mL Precellys hard tissue MK28-R tubes prefilled with 2.8 mm stainless steel grinding beads were used in a Precellys homogenizer (Bertin Corp., Rockville MD) set to 8500 rpm. Fifteen 30-s cycles were required to fully break up juvenile cuticles, but ten cycles were sufficient for adults. A 2-min rest on ice between homogenization cycles was used to cool the tubes. Mixed with 800 µL of acetonitrile were 100µL of worm homogenates, which were vortexed at 3,000 rpm for 30 s and then centrifuged for 10 min at 12,000 rpm to precipitate protein. Prior to analysis, 50 µL of the sample plus 50 µL of the internal standard were transferred to liquid chromatography amber glass vials with an integrated micro-insert for UHPLC-ESI-MS/MS analysis.

An overview of the life stage, exposure duration, and assessment period for each of the experiments in this study is provided in Figure 1.

Emulsions were prepared as described in Section 2.3 to produce 10× dosing solutions that comprised 5% sesame oil in a mixture of 1% Tween 80 in homogenized non-fat cows’ milk (1%Tw80). We mixed 20µL of vehicle control emulsion (VCe), purified CBD emulsion, or hemp extract emulsion with 1,980 µL water for zeta potential and particle size assessments.

Zeta potential analysis detected mean electrostatic charge on emulsion particles for 1%Tw80, VCe, CBD, and hemp extract (Extract) ranging from −20 to −30 mV (Figure 2A). BI-ZR5, an internal control for negatively charged particles, registered within the reference range. Dynamic light scattering indicated that the mean size of particles in the 1%Tw80 control was approximately 200 nm, and the emulsified oil mixtures did not differ significantly from this value (Figure 2B). Since CeHM is 20% non-fat cow’s milk, these data are consistent with the C. elegans responding to the emulsion particles in the same way as to their typical food source.

Diets high in fat are associated with oxidative stress in mice, rats, and humans (Le Lay et al., 2014; Ding et al., 2023; Noeman et al., 2011; Tan and Norhaizan, 2019). Expression from the promoter of the C. elegans glutathione S-transferase 4 (gst-4) gene, attached to the coding sequence for green fluorescent protein (GFP) in the transgenic strain CL2166 is used as a bioreporter of oxidative stress response (OxStrR) (Link and Johnson, 2002; Blackwell et al., 2015). To test for OxStrR, CL2166 worms were exposed from the beginning of the young adult stage, which began approximately 65 h post initial feeding after hatching. The vehicle control nanoemulsion (VCe) exposed the C. elegans to 0.5% sesame oil, a high-fat diet for this model. With exposure to VCe for 24 h, OxStrR transgene expression was increased three-fold over the matched oil-free vehicle control exposure of 0.1% Tween 80 in cows’ milk (0.1%Tw80) (Figures 3A,B).

CBD has been shown to reduce markers of inflammatory and oxidative processes in human cell cultures and laboratory mammals (Atalay et al., 2019). In C. elegans exposed to purified CBD or hemp extract for 24 h from the onset of adulthood, OxStrR gene expression was dose-dependently reduced (Figure 3, dotted bars). Compared to the VCe control, purified CBD reduced the HFD-induced OxStrR response to approximately 50% over the oil-free 0.1%Tw80 control at 100 μg/mL (Figure 3A), while an equivalent concentration of CBD in hemp extract similarly reduced reporter expression at 50 μg/mL (Figure 3B). Hemp extract at 200 μg/mL CBD and above reduced HFD-induced reporter expression to within 25% of the oil-free control, while this occurred for purified CBD only at 500 μg/mL and above.

In microfluidic laser instrument measurements, time-of-flight (TOF) is an indication of body size. Purified CBD significantly reduced TOF (Figure 3A, gray striped bars) at exposures of 400 μg/mL and above, consistent with acute toxicity at high CBD concentrations. Although purified CBD reduced OxStrR biomarker expression to below the 0.1%Tw80 oil-free control at 750 μg/mL and above, this reduction in transcription may be due to toxicity, as indicated by reduced TOF, rather than reduced oxidative stress. In contrast, hemp extract significantly reduced TOF only at the highest tested concentration. Note that in this set of experiments, the highest achieved CBD concentration in hemp extract was 960 μg/mL, not 1,000 μg/mL as with purified CBD.

Wild-type adult C. elegans were assessed for acute effects on spontaneous locomotor activity with exposure to emulsified CBD or hemp extract containing matched concentrations of CBD. The data presented represent experiments with five independently generated cohorts, using three independent emulsion dosing preparations. One of the five experiments from the hemp extract dataset was excluded due to not meeting all wAAT acceptance criteria for negative controls as described in Hunt et al. (2023).

In rodents, high-fat diets can reduce locomotor activity (Arika et al., 2019; Han et al., 2020). A comparison of the 0.5% sesame oil vehicle control emulsion (VCe) versus the oil-free (0.1%Tw80) controls from both datasets relative to their well-matched baseline activity prior to dosing demonstrates that VCe significantly reduced motor activity by a mean of 19% relative to 0.1%Tw80 across the 18-h test period and that the 95% confidence intervals around the means no longer overlap after 2 h of exposure (Figure 4A).

In some, but not all, rodent studies, oral CBD exposure can produce mild hyperlocomotion (Britch et al., 2017; Hlozek et al., 2017). Significant increases in locomotor activity above VCe were detected after exposures to 200, 400, and 750 μg/mL CBD, either purified or in hemp extract (Figure 4B, ^ symbol), consistent with CBD’s amelioration of the locomotion effects of a high-fat diet. Other concentrations also increased activity above VCe, but the variability (as indicated by the 95% confidence intervals shown in parentheses in Figure 4B) was too large for significance. Populations exposed to 400 or 750 μg/mL CBD from either the purified chemical or in hemp extract significantly differed from each other (Figure 4B, # symbol), with the hemp extract closer to the oil-free control. Purified CBD at 200 μg/mL to 500 μg/mL restored spontaneous locomotor activity to 0.1%Tw80 levels within approximately 8–12 h of exposure (Figure 4C), while none of the hemp exposures tested were significantly different from the oil-free 0.1%Tw80 control over the 18 h assessment period (Figures 4B,D). Together, these data indicate that purified CBD was less effective than CBD in hemp extract in counteracting the hypoactivity effects of the sesame oil emulsion in adult C. elegans.

Purified CBD at 750 μg/mL and 1,000 μg/mL also increased motor activity above VCe but to a lesser extent than lower concentrations of purified CBD or matched concentrations of CBD in hemp extract (Figures 4B,C). These data are consistent with a U-shaped activity curve and a reduction in benefit at 750 μg/mL and above for purified CBD. Acute toxicity, as measured by a reduction in motor activity to less than 60% of control, as previously observed with mercury and arsenic (Hunt et al., 2023), did not occur at any tested CBD concentration in either preparation.

The C. elegans progeny-to-adult ratio assay measures the number of progeny relative to the number of adults in control and exposed populations, and it also provides information on microfluidic/laser technology morphometry as time-of-flight (TOF), a measure of size. Age-synchronized C. elegans were continuously exposed from the onset of reproductive maturity in the parental cohort. Replicate plates were assessed at 2-, 3-, and 4-days post-dosing, equivalent to days 5–7 post L1 feeding (pL1f) of the parental population. Initial testing was also conducted on parental day 8 pL1f, but the control parental and progeny populations could no longer be separated by gating at that point.

For populations exposed to purified CBD at 400 μg/mL and above, progeny output relative to the matched vehicle control emulsion (VCe) was significantly reduced in a dose-dependent manner, and these reductions were relatively consistent over the 3-day assessment period (Figure 5A). However, for the matched CBD concentrations in the hemp extract, only 750 μg/mL and 960 μg/mL significantly reduced progeny output, and only on days 2 and 3 post-dosing (d2pD and d3pD) (Figure 5B). Based on the reproductive effects of individual concentrations of purified CBD and hemp extract normalized to CBD concentration, there was a marked difference between the two, especially at the three highest tested concentrations. For example, with exposures of 750 μg/mL purified CBD, the ratio of progeny to adults was reduced by 71% on d2pD, and 62% on d3pD and d4pD relative to VCe controls. At 750 μg/mL CBD in hemp extract, the ratio of progeny to adults was reduced by 32%–33% on d2pD and d3pD, and a non-significant mean downward trend of only 7% on d4pD.

Each independent experiment for CBD or hemp extract was performed with three replicates (one for each day of assessment) of two plates, one for low concentrations and the other for high concentrations plus the oil-free 0.1%Tw80 and 5-fluoro-2′-deoxyuridine (FUdR) controls, with each plate having a VCe control. The difference in progeny output between the VCe controls on the two plates was not significantly different at any assessed time point. Progeny output for the 0.1%Tw80 control was reduced by a mean of 17% only for the CBD set of experiments, and only for d4pD. This may have been due to slight overcrowding of C. elegans by the fourth day of continuous exposure and consequent reduction in nutrient resources. FUdR at 5 μg/mL was used as a positive control for reducing reproductive output. In FUdR-exposed populations, few or no progeny were observed by microscopy or counted by the COPAS (Figures 5A,B). In contrast to FUdR’s effects on reproduction, adults exposed to FUdR had normal body size as measured by TOF (Figures 5C,D).

On d2pD, purified CBD was associated with mean adult body size reductions of 10%, 22%, and 30% at 500 μg/mL, 750 μg/mL, and 1,000 μg/mL, respectively (Figure 5C). On d3pD, smaller mean changes in adult body size became statistically significant at lower concentrations, with 5% reductions in TOF at 300 μg/mL and 400 μg/mL (Figure 5C); however, the overall dose-dependent trend did not change for purified CBD over the 3 days of assessment. In contrast, CBD in hemp extract only reduced adult body size by 5%, 11%, and 14% at 500 μg/mL, 750 μg/mL, and 960 μg/mL CBD, respectively, and only on d2pD (Figure 5D). This more modest reduction in adult body size, and only at the earlier stage of chronic hemp extract exposure, is consistent both with a) progeny ratio data showing recovery of progeny output by d4pD (Figure 5B) and with b) significant reductions in adult body size in the oxidative stress reporter strain only at the highest 960 μg/mL CBD in hemp extract exposure (Figure 3B).

The mean juvenile population body size was reduced dose-dependently both with purified CBD and with matched concentrations of CBD in hemp extract, although again the lowest observed adverse effect levels (LOAELs) were lower for purified CBD relative to CBD in hemp extract (Figures 5E,F). Additionally, and in contrast to adults, there was a significant downward shift in mean juvenile body size with continued population exposure, suggesting possible germ cell-mediated developmental toxicity in later progeny and/or embryonic effects. By d4pD, the mean juvenile population body size LOAEL was 200 μg/mL for purified CBD (Figure 5E) or 400 μg/mL for CBD in hemp extract (Figure 5F), with reductions of 12% for both. At 750 μg/mL continuous purified CBD exposure, mean juvenile TOF values were reduced by 27% on 2dpD and d3pD, and by 46% on d4pD. At 750 μg/mL CBD in hemp extract, mean juvenile TOF values were reduced by 11%, 22%, and 29% on days 2, 3, and 4 post-dosing, respectively. These data are consistent with the cumulative effects on progeny with continuous chronic exposure from the time of parental reproductive maturity.

Under control conditions, C. elegans hermaphrodites are reproductively active for approximately 6 days after reaching the adult stage (Johnson, 1987). With testing through 4 days of continuous exposure beginning at the onset of reproductive maturity, the progeny-ratio-assay-identified C. elegans No Observed Adverse Effect Levels (NOAELs) for purified CBD to be 300 μg/mL for reproductive output, 200 μg/mL for adult toxicity as measured by reduced mean body size, and 100 μg/mL for juvenile toxicity as measured by reduced mean body size. NOAELs identified for CBD in hemp extract were 500 μg/mL for reproductive output, 400 μg/mL for adult toxicity, and 300 μg/mL for juvenile toxicity.

The wDAT uses an infrared beam locomotion tracking device to monitor both developmental timing and stage-specific locomotor activity in wells containing approximately 900 worms each, with three replicate wells per plate. C. elegans develop through four larval stages, designated “L1–4”, which can be monitored with the motion detection equipment due to lethargus, the brief period of reduced activity during cuticle molting between developmental stages (Figure 6A). This can be used to assess chemical effects on growth and behavior and also gives an indication of the synchrony of developmental timing within the population. The lack of decline in population locomotion after an activity peak is consistent with the microscopically observed asynchronous maturity in an exposed cohort. Peak time and height can be estimated for minor losses in synchronous cohort development, as indicated by the orange arrows pointing to the estimated L4 activity peaks for 500 μg/mL and 750 μg/mL CBD in Figure 6A. The orange “X” in Figure 6A indicates an example of a growth curve for a population that is developing too asynchronously to estimate an activity peak.

As originally designed, the wDAT evaluates the effects of continuous exposure from the onset of feeding after hatching through to the adult stage. To evaluate the reversibility of developmental delays and motility changes caused by CBD, an “early-only” exposure paradigm was also used. For the early-only exposures, approximately 3,000 worms in single wells were exposed for 24.5 ± 0.5 h from initial L1 feeding, followed by washing, re-feeding, and transfer to three wells for further monitoring (Figure 6B). Representative individual continuous vs early-only exposure wDAT experiments demonstrate that continuous exposure to purified CBD caused both hypoactivity and developmental delay (Figure 6A), while early-only exposures were associated with later developmental delays but not later hypoactivity (Figure 6B). Initial range-finding experiments found small and inconsistent effects at CBD concentrations of 300 μg/mL and below; therefore, the subsequent experiments used 300 μg/mL as the minimum developmental exposure.

In a set of four independent continuous exposure experiments, purified CBD induced delays of ∼5%–40% in a dose-response manner from 400 μg/mL to 1,000 μg/mL (Figure 6C). CBD-induced delays were relatively consistent across all four larval stages for most exposures, with the exception that the 5% delay to mid-L1 reached statistical significance at 300 μg/mL CBD, and delays to L2 tended to be both greater than delays to the other stages and more variable (Figure 6C). Higher concentrations of purified CBD induced some but not total, loss of synchronous developmental timing within the population; at 1,000 μg/mL CBD, an L4 peak had to be estimated in three out of four total experiments (Figure 6C, indicated by an orange “!3”). The effects on locomotion were quite different for different stages, with significant hypoactivity at the L1 and L4 stages with 400 μg/mL to 1,000 μg/mL purified CBD, L3 hypoactivity only at the highest concentration, and hyperactivity at L2 with 500 μg/mL and 750 μg/mL CBD (Figure 6D). The raw data for this set of experiments is provided in Supplementary Data Sheet 1.

In a second set of wDAT experiments, continuous and early-only exposures for CBD and hemp extract were conducted simultaneously in order to distinguish reversible from irreversible developmental effects. For controls at 20°C in CeHM, the L1 stage lasts approximately 19 h and is entirely within the 24 h exposure period. Therefore, as expected, the L1 data from the continuous and early-only exposure data sets were not statistically different from each other (Figure 7A). For hemp extract exposures normalized to CBD concentration within the extract, L1, and L2 delays were similar to those seen with purified CBD (Figure 7B). As shown in Figure 6B, L2 locomotor activity was disrupted by washing and re-feeding; therefore, for L2, only the continuous exposure effects were evaluated. In this second set of experiments, purified CBD was not associated with significant L2 hyperactivity, but hemp extract was associated with hyperactivity at L2 (Figures 7C,D). The raw data for this set of experiments is provided in Supplemental Data Sheet 2.

At 20°C, control L3 activity peaks usually occur 38–40 h post initial L1 feeding, more than 12 h after removing the test article for early-only exposures. Not only were later delays not reduced by CBD removal, but there was a trend toward increased delays to L3 for early-only exposure populations (Figures 7E,F, striped bars) relative to continuous exposures (Figures 7E,F, solid bars). This trend achieved statistical significance at 500 μg/mL purified CBD, with L3 delays of 9% for continuous exposure but 17% with early-only exposure (Figure 7E, orange asterisk) and at 750 μg/mL CBD in hemp extract (Figure 7F, orange asterisk). In contrast to these irreversible effects on developmental delay, early-only CBD or hemp extract exposure was not associated with the later hypoactivity at L3 and L4 (Figures 7G,H), indicating that hypoactivity is a reversible effect.

In juveniles, the effects of the vehicle control emulsion relative to the oil-free control were inconsistent, with increased levels of activity with 0.1%Tw80 relative to VCe in some sets of experiments (Figures 6D, 7D) but not in others (Figures 7B,F,H).

Adult and L1 C. elegans were exposed to CBD or hemp extract in emulsions for 24 h, then washed, homogenized, and assessed as described in Section 2.8 in four independent experiments. Mean internal CBD concentrations were 0.15, 0.22, and 0.29 ng CBD per juvenile C. elegans exposed to 300, 500, and 750 μg/mL purified CBD, respectively (Figure 8A). Although not statistically different from the purified CBD data set, the values were lower for hemp extract with 0.12, 0.13, and 0.19 ng CBD per juvenile worm exposed to hemp extract containing 300, 500, and 750 μg/mL CBD, respectively. Mean internal CBD concentrations were 1.47, 2.30, and 2.32 ng CBD per adult C. elegans exposed to 300, 500, and 750 μg/mL CBD, respectively (Figure 8A). For hemp extract, the values were 1.27, 2.00, and 2.85 ng CBD per adult worm for adults exposed for 24 h to hemp extract containing 300, 500, and 750 μg/mL CBD, respectively. Although the mean values indicate a trend toward increasing internal CBD concentrations with increasing exposure concentrations, only the adult worms exposed to 300 vs. 750 μg/mL CBD in hemp extract were statistically different from each other with a Student’s t-test p-value of <0.05. CBD was not detected in worms exposed to the vehicle control.

The volume of a juvenile C. elegans 24 h after L1 feeding (24hpL1f) has been calculated to be approximately 0.35 nL, and the volume of an adult to be approximately 6 nL (Knight et al., 2002). Therefore, a pellet containing approximately 10,000 adult C. elegans should have a volume of approximately 60µL, which is consistent with our observations. The density of well-fed L1 through L3 C. elegans has been measured to be approximately 1.091 g/mL and adult density to be approximately 1.074 g/mL 24 h after the L4 adult molt (Reina et al., 2013). Taking these published values together, the approximate mass/worm would be ∼0.38 µg for 24hpL1f C. elegans and ∼6.44 µg for adults. Dividing the nanogram CBD-per-worm values above by the calculated mass per worm, tissue concentrations ranged between 200–and 770 ng CBD per mg C. elegans tissue, with the highest tissue concentrations at each exposure being in juveniles exposed to purified CBD (Figure 8B). Although there was a trend toward higher CBD levels in purified CBD-exposed juvenile tissues, the juvenile versus adult and purified CBD versus hemp extract-exposed tissues were not statistically different at any tested concentration (Figure 8B).