Assessment of the effects of cannabidiol and a CBD-rich hemp extract in Caenorhabditis elegans

Abstract

Consumer use of cannabidiol (CBD) is growing, but there are still data gaps regarding its possible adverse effects on reproduction and development. Multiple pathways and signaling cascades involved in organismal development and neuronal function, including endocannabinoid synthesis and signaling systems, are well conserved across phyla, suggesting that Caenorhabditis elegans can model the in vivo effects of exogenous cannabinoids. The effects in C. elegans on oxidative stress response (OxStrR), developmental timing, juvenile and adult spontaneous locomotor activity, reproductive output, and organismal CBD concentrations were assessed after exposure to purified CBD or a hemp extract suspended in 0.5% sesame oil emulsions. In C. elegans, this emulsion vehicle is equivalent to a high-fat diet (HFD). As in mammals, HFD was associated with oxidative-stress-related gene expression in C. elegans adults. CBD reduced HFD-induced OxStrR in transgenic adults and counteracted the hypoactivity observed in HFD-exposed wild-type adults. In C. elegans exposed to CBD from the onset of feeding, delays in later milestone acquisition were irreversible, while later juvenile locomotor activity effects were reversible after the removal of CBD exposure. CBD-induced reductions in mean juvenile population body size were cumulative when chronic exposures were initiated at parental reproductive maturity. Purified CBD was slightly more toxic than matched concentrations of CBD in hemp extract for all tested endpoints, and both were more toxic to juveniles than to adults. Dosimetry indicated that all adverse effect levels observed in C. elegans far exceeded recommended CBD dosages for humans.

Keywords: C. elegans; alternative in vivo toxicity test model; cannabidiol; chronic exposure effects; hemp extract; irreversible developmental effect.

FIGURE 1

Diagram of experimental exposure and assessment periods. The time course of C. elegans life stages is shown from the first feeding after hatching (feed) through the four larval stages (L1–L4), young adult (yA, defined as producing oocytes but not yet having fertilized eggs in the uterus), and adult (A, at least one internal fertilized egg) stages. The oxidative stress response (OxStrR) was assessed using CL2166 C. elegans exposed for 24 h from the young adult stage, followed by assessment of fluorescence (a measure of gene expression) and time-of-flight (TOF, a measure of size). The worm Adult Activity Test (wAAT) assessed contraceptive (FUdR)-exposed adult wild-type C. elegans for spontaneous locomotor activity changes. The wAAT assay included a 2-h baseline acquisition prior to exposure, followed by exposure and tracking for 18 h. The worm Development and Activity Test (wDAT) tracked the timing and spontaneous locomotor activity of wild-type C. elegans. For continuous wDAT exposure, synchronous cohorts are exposed within 30 min of first feeding through to the adult stage. For early-only wDAT exposures, cohorts were exposed for the first 24 h following the first feeding, then washed, re-fed, and put back into the tracker for assessment through to the adult stage. For whole-worm tissue concentration assessments, 24 h exposures were begun either at first feeding or at the yA stage, using wild-type C. elegans. After exposure, worms were washed, homogenized, extracted, and frozen for later (∼) analysis. The developmental timing of C. elegans depends on temperature. Progeny ratio experiments were conducted using PD4251 C. elegans incubated at 19°C so that all experimental steps could be accomplished during the daytime. Progeny ratio exposures were continuous, beginning from the young adult stage, and replicate plates were used for analyses on days 2, 3, and 4 post-dosing.

FIGURE 2

Emulsion particle charge and diameter measurements. 10× dosing emulsions were prepared using 5% sesame oil in a blend of 1% Tween 80 in non-fat cow’s milk (1%Tw80). (A) Mean emulsion particle charge, using BI-ZR5 as an internal reference for negatively charged particles. (B) Mean particle size, using 92-nm latex beads as an internal reference (92 nm std.). The number of independent experiments is listed in bold at the base of each column. Error bars indicate the standard deviation over the number of assessments listed.

FIGURE 3

Oxidative stress bioreporter (OxStrR) gene expression and biometric analysis. GFP fluorescence and time-of-flight (TOF, a measure of body size) were measured with a COPAS microfluidic laser instrument using the transgenic C. elegans strain CL2166, a bioreporter for changes in oxidative stress response gene expression (OxStrR, dotted bars). (A, B) CBD, purified or in hemp extract, reduced OxStrR in a dose-dependent manner. (A) Exposure to purified CBD for 24 h from the onset of adulthood significantly reduced TOF (striped bars) at exposures of 400 μg/mL and above. (B) Hemp extract at matched CBD concentrations significantly reduced TOF only at 960 μg/mL. The number of independent experiments is listed in bold at the base of each column, and the error bars indicate the standard deviation among the listed number of experiments (note that the variance for TOF was very small, making the error bars appear as a single line). Bars indicate changes normalized in each experiment to the 0.1%Tw80 control. Student’s t-test p-values from a minimum of four independent experiments of (*) ≤ 0.05 and (#) ≤ 0.005 were considered statistically meaningful. Minimum mean changes of 5% or 10% from control were considered biologically meaningful for TOF and fluorescence, respectively. Symbols in black indicate significant differences from the VCe control for GFP bioreporter expression, while red asterisks indicate significant differences from the 0.1%Tw80 control for TOF.

FIGURE 4

Spontaneous locomotor activity in synchronized adult populations. Mean population locomotor activity in wild-type adult C. elegans was measured with an infrared beam interruption device and analyzed with the web tool https://waat.galaxytrakr.org. (A) Graph of the mean of all oil-free control (0.1%Tw80, upper blue line) and vehicle control emulsion (VCe, lower pink line) activity levels over the course of the 18-h test for both CBD and hemp extract data sets, and the shaded areas indicate the 95% confidence band around the mean values. (B) Data over the 18-h test period using the model-predicted minimum activity as the determinant profiler is shown, with the model-predicted minimum value in bold and the confidence interval in parentheses. Statistically significant minimum activity greater than VCe is indicated by a ^ symbol, while # indicates concentrations where purified CBD was significantly different from CBD in hemp extracts. (C) Graph of the mean effects of purified CBD on activity levels relative to 0.1%Tw80. (D) Graph of the mean effects of CBD in hemp extract effects on activity levels relative to 0.1%Tw80.

FIGURE 5

Progeny-to-adult ratio and biometric analysis. Assessment on days 2, 3, and 4 post-dosing of the ratio of progeny to adults plus morphometry in populations continuously exposed from the onset of reproductive maturity in the parental population. This assay uses PD4251, a C. elegans strain that glows bright green upon laser excitation, to allow for the separation of the smallest progeny from the background noise generated by the nutrient media. (A, B) Ratio of progeny to adults for purified CBD, and for hemp extract concentrations matched for CBD content. (C, D) Mean adult population time-of-flight (TOF, an indicator of size). (E, F) Mean TOF of the progeny population. Changes were considered meaningful if the mean difference (three independent experiments minimum) from vehicle emulsion (VCe) control was ≥5% and values normalized to the plate-matched controls had Student’s t-test p-values of <0.05 (*) or <0.005 (#).

FIGURE 6

Worm Development and Activity Test (wDAT). (A) A single continuous exposure wDAT experiment is graphed. Activity peaks for the four larval stages are marked L1–4. “VCe L” and “VCe R” refer to the mean activity in vehicle control emulsion wells in simultaneously run left and right plates, respectively. Orange arrows indicate estimated peak times and heights. The orange “X” indicates a high concentration curve for which development within the population was not sufficiently synchronized to estimate an L4 peak. (B) A single early-only exposure wDAT experiment is graphed. The washing and re-feeding period is indicated by a break in the data at 25 h. Developmental delays, as indicated by a right shift of the peaks, remained after CBD was removed, but the hypoactivity, as indicated by reduced peak height in (A), was not observed. (C, D) Data from a set of four independent continuous exposure CBD experiments are graphed. “VCe L-R” indicates the difference between control wells in plates run side-by-side, indicating experimental variability. Orange “!” symbols followed by a number indicating how many experiments for which L4 peak time needed to be estimated, consistent with loss of synchronous development at higher concentrations of CBD. For each experiment, mean values across three wells were normalized to the plate-matched vehicle emulsion (VCe) control wells. Changes were considered biologically meaningful if the mean difference from the vehicle emulsion (VCe) control was ≥5% for delay and ≥10% for activity. Significant differences in Student’s t-test p-values of <0.05 or <0.005 are indicated by * or #, respectively.

FIGURE 7

Assessment of the reversibility of adverse developmental outcomes. Effects of purified CBD and hemp extract at each of the four larval stages from continuous and early-only exposures are shown as solid and striped bars, respectively. Note that due to the timing of the washing steps in the early-only exposures, L2 data were evaluated for continuous exposures only. Effects of (A) purified CBD and (B) hemp extract on the time to reach mid-L1 and mid-L2. Effects of (C) purified CBD and (D) hemp extract on spontaneous locomotor activity at L1 and L2. (E–H) Effects of CBD and hemp extract on L3 and L4 developmental timing and activity. An orange asterisk or “#” symbol indicates where there was a significant difference between the continuous and early-only exposures at the same concentration. This figure represents five independently derived cohorts exposed to three sets of independently prepared emulsions for the early-only data. However, controls from two of the five continuous exposure experiments graphed in E–L did not meet acceptance criteria, so data were based on only three experiments for this set, which was reflected in higher p-values and lower confidence. For each experiment, the mean values across three replicate wells were normalized to the plate-matched vehicle emulsion (VCe) controls. VCe ϴ indicates control change values in the left plate, which are defined as zero, and VCe L-R indicates the difference between control wells in the left and right plates, giving an internal measure of experimental variability. Changes were considered meaningful if the mean difference from the plate-matched vehicle emulsion (VCe) control was ≥5% for delay and ≥10% for activity. Significant differences are indicated as Student’s t-test p-values of <0.05 (*) or <0.005 (#).

FIGURE 8

Assessment of tissue concentrations of CBD in C. elegans. (A) CBD concentrations in C. elegans exposed and analyzed as described in Section 2.8. Bars indicate the mean value of four independent experiments, and error bars indicate the standard deviations. (B) Concentration in ng CBD per mg C. elegans tissue, calculated as described in Section 3.6.