Introduction

Despite the growing popularity and availability of various cannabis products, smoke-inhalation remains the preferred method of cannabis consumption amongst pregnant people1. Although preclinical and clinical studies have linked prenatal cannabis use and its constituents to adverse outcomes, such as low birth weight, preterm birth, fetal growth restriction and poor neonatal complications, there is still conflicting evidence regarding these outcomes2,3,4,5,6. Additionally, there remains uncertainty in the literature largely because human studies are often confounded by polysubstance use, cannabis potency, variations in cannabis cultivars, and the frequency/duration of exposure7,8,9.

Cannabis contains over 500 distinct chemical compounds including the highly abundant cannabinoids extracted for human use,Δ9-tetrahydrocannabinol (Δ9-THC) and cannabidiol (CBD)10,11. Like tobacco smoke, cannabis smoke also contains a complex mixture of combustion by-products such as polycyclic aromatic hydrocarbons (PAHs), nitrogen oxides, particulate matter and tar12. Most pre-clinical studies focus on the effects of individual injections of THC or CBD5,6,13,14,15,16, which may overlook potential complex effects of the other constituents routinely found in Cannabis, including the diverse group of phytocannabinoids and terpenes17 and the combustion by-products.

The adverse outcomes that have been reported with cannabis use during pregnancy such as preterm birth and low birth weight have been linked to placental dysfunction18. The chorionic villi of the placenta are the primary sites of materno-fetal nutrient, metabolic and hormonal exchange19. These structures are covered by a layer of multi-nucleated syncytiotrophoblast (ST), which continually differentiate and fuse throughout pregnancy from primitive underlying cytotrophoblast (CT) to establish and maintain placental function20. Placentae are rich in mitochondria, which serve as key metabolic sensors that support the energy-intensive demands of cellular dynamic processes during pregnancy, including cell commitment and differentiation21. Previously, our research group has demonstrated that exposure to Δ9-THC and CBD in vitro disrupts mitochondrial electron chain function as well as trophoblast fusion in BeWo cells22,23. Similarly, animal models of pregnancy with Δ9-THC and/or CBD have also shown placental defects5,6. However, recent work has shown differential effects on placental and fetal cytokine and chemokine profiles, as well as on maternal and fetal outcomes when comparing cannabis smoke to injectable forms in a mouse model of pregnancy, suggesting the route of exposure may distinctly impact placental-related outcomes24,25. Determining the effects of cannabis smoke on trophoblast syncytialization and overall placental health is crucial in enhancing our understanding of cannabis-induced pregnancy complications whilst considering real-world cannabis usage patterns. Therefore, we hypothesized that THC-dominant cannabis smoke extract (CaSE) would impair trophoblast differentiation and mitochondrial respiration in BeWo b30 cells. We demonstrate that while both CaSE and Δ9-THC disrupt mitochondrial function and trophoblast differentiation, the mechanism of action may differ between the two exposures. Taken together, these findings underscore the importance of distinguishing between the effects of single components like Δ9-THC, and the complex mixture of constituents present in cannabis smoke when evaluating risks to placental development and reproductive health.

Results

Cannabis smoke exposure induces CYP1A1, and Δ9-THC is detectable in CaSE-treated cells

To measure Δ9-THC and CBD concentrations in our CaSE, we performed LC-MS/MS on CaSE media before and after 0.22 μm filtration, as well as on cultured cell supernatants collected 48 h post-treatment. Quantification confirmed that following filtration, there was a 48.7 ± 13.0% (mean ± s.d., n = 3) reduction in Δ9-THC levels in the media. The post-filtration media contained an equivalent of 224 µM THC (100% CaSE). In sample supernatant of cells treated for 48 h with CaSE, Δ9-THC concentrations were found to be 1.74 µM and 4.30 µM in the 2.5% and 5.0% CaSE, respectively (Fig. 1a, p < 0.0001) alongside no substantial detection of CBD levels (Fig. 1b) compared to the vehicle. Representative chromatograms of pre-filtered and post-filtered media as well as sample supernatant can be found in Supplemental Figure S2 (a-c). To assess the extent to which CaSE stimulated genes involved in metabolic and bioactivation pathways, we evaluated gene expression levels involved in the aryl hydrocarbon receptor (AhR) signaling cascade. Gene expression levels of CYP1A1 (cytochrome P450 family 1 subfamily A member 1), a key enzyme involved in metabolizing PAHs, were upregulated by 900-fold and 1600-fold in 2.5% and 5.0% CaSE, respectively, compared to vehicle (Fig. 1c, p < 0.0001). In contrast, Δ9-THC alone did not result in a significant change in the expression of CYP1A1 (Fig. 1c). LC-MS/MS quantification of cannabinoid metabolites alongside CYP1A1 induction, collectively validated the extent of trophoblast exposure to cannabis smoke constituents.

Fig. 1
Fig. 1The alternative text for this image may have been generated using AI.
Full size image

CaSE induces CYP1A1 expression and contains detectable THC concentrations. BeWo b30 cells were treated with vehicle (0.1% methanol), 10 µM THC, 2.5% or 5.0% CaSE for 48 h during differentiation. (a, b) Media supernatant was collected 48 h following exposure to CaSE and subjected to LS-MS/MS for metabolite quantification (THC, CBD). The limits of quantitation (LOQ) were 25.3 pg/µL for THC and 23.5 pg/µL for CBD. Transcript expression (c) in each treatment group was normalized to the GEOMEAN of 18 S and β-Actin, then compared to the gene in the control group, assessed using RT-qPCR. Significant differences were determined by a one-way ANOVA and corrected by a post-hoc Tukey test. Data presented as mean ± SEM (n ≥ 5). Different letters denote significant differences of p < 0.05.

Cannabis smoke exposure reduces cellular viability along with biochemical markers of trophoblast fusion and differentiation

Treatment of BeWo b30 cells with increasing concentrations of CaSE resulted in a sharp decrease in cellular viability at CaSE concentrations greater than 10% (Supplementary Fig. S1, p < 0.0001). At lower dosages of CaSE treatment, the cellular viability remained above 80%. Therefore, 10% CaSE was included in select functional assays to demonstrate the threshold of exposure-induced cytotoxic and metabolic responses but was omitted from molecular analyses (qPCR and Western blotting) due to limited cell viability. Further evaluations were carried out for markers of trophoblast fusion and differentiation using concentrations of 2.5% and 5% CaSE as well as 10 µM Δ9-THC; a concentration which we have previously shown to be non-cytotoxic in this cell line22. The impact of Δ9-THC and CaSE on trophoblast differentiation was assessed by quantifying the mRNA levels of human choriogonadotropin hormone (hCG), CGα (chorionic gonadotropin alpha), and CGβ (chorionic gonadotropin beta), as well as the total protein levels of hCG. Treatment with Δ9-THC exhibited a significant reduction in CGβ transcript levels (p < 0.0001), and no significant changes to hCG protein levels (Fig. 2a-c). However, 2.5% and 5.0% CaSE-treated cells showed a significant reduction in both CGα (Fig. 2a, p < 0.0001) and CGβ (Fig. 2b, p < 0.0001). In comparison to THC treatment, 2.5% and 5.0% CaSE treatments resulted in significant reductions in hCG protein content (Fig. 2c, p < 0.05). Given cell differentiation is inversely related to proliferative capacity, we assessed gene levels of a marker of cytotrophoblast proliferation, KI67 (antigen Kiel 67), which was found to be decreased in the THC group (Fig. 2d, p < 0.0001) and increased in the 5.0% CaSE group (Fig. 2d, p < 0.05). As differentiation relies upon appropriate cell-cell fusion, we assessed cell fusion marker ERVW-1 (endogenous retrovirus group W, member 1) and found mRNA levels decreased by treatments of THC (Fig. 2e, p < 0.01), and in both 2.5% and 5.0% CaSE-treated cells (Fig. 2d, p < 0.001).

Fig. 2
Fig. 2The alternative text for this image may have been generated using AI.
Full size image

CaSE impairs trophoblast fusion and differentiation. BeWo b30 cells were treated with vehicle (0.1% methanol), 10 µM THC, 2.5% or 5.0% CaSE for 48 h during differentiation. Transcript expression (a, b, d, e) in each treatment group was normalized to the GEOMEAN of 18 S and β-Actin. Cell lysates (25 µg/lane) were loaded and seperated on a 10% SDS-PAGE and the content of hCG was quantified using a rabbit polyclonal anti-hCG antibody. Summary histogram of relative protein levels of (c) hCG (n = 4) were assessed using Western blotting and normalized to the stain-free image. Cell lysates (25 µg/lane) were loaded on a 10% gel and SDS-PAGE and quantified using a rabbit polyclonal anti-hCG (DAKO, GA508) antibody. Significant differences were determined by a one-way ANOVA and corrected by a post-hoc Tukey test. Data presented as mean ± SEM (n ≥ 5). Different letters denote significant differences of p < 0.05.

Cannabis smoke exposure induces markers of cellular and oxidative stress, along with increased ROS levels

We investigated the cellular and mitochondrial stress related biochemical changes following cannabis smoke exposure by quantifying mRNA levels of HSP60 (heat shock protein 60) and HSP70 (heat shock protein 70), respectively. Cells treated with Δ9-THC for 48 h exhibited no changes in HSP60. In contrast, HSP60 expression was significantly upregulated following 2.5% CaSE and 5.0% CaSE treatments (Fig. 3a, p < 0.001). Conversely, cells treated with THC and both 2.5% and 5.0% CaSE for 48 h exhibited a downregulation in HSP70 mRNA levels (Fig. 3b, p < 0.05). Additionally, we assessed mRNA levels of antioxidant enzymes, superoxide dismutase 1 and 2, SOD1 and SOD2. While transcript levels of SOD1 were unchanged in Δ9-THC-treated cells, there was an upregulation in 2.5% and 5.0% CaSE-treated cells (Fig. 3c, p < 0.0001). Furthermore, SOD2 gene expression levels were increased in THC-treated cells (Fig. 3d, p < 0.05) and in 5.0% CaSE treated cells (Fig. 3d, p < 0.01). Due to the apparent alterations in superoxide dismutase expression, we quantified the intracellular levels of reactive oxygen species (ROS). We observed ROS levels were significantly increased with all CaSE treatments (Fig. 3e, p < 0.001); however, there were no substantial changes to ROS levels with Δ9-THC treatments. We then measured mitochondrial membrane potential using the JC-1 assay. Following treatment with 5.0% and 10% CaSE there was a reduction in membrane potential (Fig. 3f, p < 0.05) relative to the vehicle.

Fig. 3
Fig. 3The alternative text for this image may have been generated using AI.
Full size image

CaSE induces ROS production and activates cellular and oxidative stress markers. BeWo b30 cells were treated with vehicle (0.1% methanol), 1 µM THC, 10 µM THC, and 1.0%, 2.5%, 5.0% or 10% CaSE for 48 h during differentiation. Transcript expression (a–d) in each treatment group was normalized to the GEOMEAN of 18 S and β-Actin, then compared to the gene in the control group, assessed using RT-qPCR. (e) The DCFDA assay was used to measure levels of reactive oxygen species; rotenone (10 nM) was used as a positive control. (f) The JC-1 assay was used to measure mitochondrial membrane potential (ΔΨm). Significant differences were determined by a one-way ANOVA and corrected by a post-hoc Tukey test. Data presented as mean ± SEM (n ≥ 5). Different letters denote significant differences of p < 0.05.

CaSE treatment results in attenuated basal respiration and ATP production

To establish the impacts of Δ9-THC and CaSE on parameters of mitochondrial respiration, we measured the oxygen consumption rate (OCR) using the Agilent Seahorse XF Mito Stress Test. We treated cells with a range of Δ9-THC and CaSE concentrations (1 µM, 10 µM, and 1%, 2.5%, 5%, 10%, respectively) for 48 h to establish dose-dependent relationships with drug exposure prior to the addition of oligomycin, FCCP (carbonyl cyanide-p-(trifluoromethoxy)phenylhydrazone) and antimycin A/rotenone. We found that 10 µM Δ9-THC and 5.0% CaSE decreased basal respiration (Fig. 4c, p < 0.05). Additionally, ATP production was also significantly reduced following 10 µM Δ9-THC and 5.0% CaSE (Fig. 4d, p < 0.001).We also observed proton leak, maximal respiratory capacity and non-mitochondrial respiration to be diminished with 5.0% and 10% CaSE (Supplementary Fig. S3).

Fig. 4
Fig. 4The alternative text for this image may have been generated using AI.
Full size image

High dose CaSE diminishes basal respiration rates and ATP production. BeWo b30 cells were treated with vehicle (0.1% methanol), 1 µM THC, 10 µM THC, 1%, 2.5%, 5.0% and 10% CaSE for 48 h during differentiation, and subjected to the mitochondrial stress test. Basal OCR tracings were obtained using the Seahorse XFe24 Analyzer using the sequential addition of oligomycin, FCCP and antimycin A/rotenone; values were normalized to total protein content via the BCA assay (µg/mL). The detection of OCR was performed with four technical replicates per experiment, for each treatment condition, and repeated three more times (n = 3 biological replicates). Group mean and SEM are displayed below (a-d). Arrows indicate the sequential addition of the respective compounds. Significant differences were determined by a one-way ANOVA and corrected by a post-hoc Tukey test. Different letters denote significant differences of p < 0.05. (a, b) Oxygen consumption rate (OCR) tracings, (c) Basal respiration, (d) ATP-linked Production.

The role of CB1 receptor-mediated signaling in facilitating the cellular effects of Δ9-THC and CaSE

To assess the impacts of CaSE and Δ9-THC on components of the endocannabinoid system (ECS), we assessed CB1 (cannabinoid receptor 1) protein levels and found no significant changes with any treatment condition (Fig. 5a). To determine whether THC- and CaSE-driven changes to markers of syncytialization were mediated via the CB1 receptor, we pre-treated our cellswith 1 µM AM281 (CB1 antagonist) for 30 min followed by subsequent drug treatments with THC or CaSE for 48 h. We found that THC-induced changes to transcript levels of CGB and ERVW-1 were rescued following CB1 receptor antagonism (Fig. 5e-f, p < 0.05). However, the effects on genes related to differentiation were not restored following CB1 receptor antagonism in CaSE-exposed cells. (Fig. 5e-f). Considering CYP1A1 gene expression levels were stimulated following CaSE treatment, we evaluated the extent of AhR (aryl hydrocarbon receptor) stimulation. Gene expression levels of AHR were unchanged following Δ9-THC treatment and CaSE exposure (Fig. 5b). However, transcript levels of AHRR (aryl hydrocarbon receptor repressor) were upregulated with 2.5% CaSE treatment (Fig. 5c, p < 0.0001).

Fig. 5
Fig. 5The alternative text for this image may have been generated using AI.
Full size image

CB1R antagonist treatment does not restore CaSE-induced disruptions to differentiation markers. BeWo b30 cells were treated with vehicle (0.1% methanol), 10 µM THC, 2.5% and 5.0% CaSE for 48 h during differentiation, and a subset of cells were pre-treated with AM281 (1 µM) for 30 min and then treated with the corresponding treatment. Summary histogram displaying the relative changes in protein levels of (a) CB1 (n = 4) were assessed using Western blotting and normalized to the stain-free image. Cell lysates (25 µg/lane) were loaded and seperated on a 4–20% SDS-PAGE and quantified using a rabbit polyclonal anti-CB1 antibody. Transcript expression (b–f) in each treatment group was normalized to the GEOMEAN of 18 S and β-Actin, then compared to the gene in the control group, assessed using RT-qPCR. Significant differences were determined by a one-way ANOVA and corrected by a post-hoc Tukey test. Data presented as mean ± SEM (n ≥ 5). Different letters denote significant differences of p < 0.05.

Discussion

To our knowledge, this study is the first to examine the effects of cannabis smoke extract (CaSE) on human placental trophoblast cells. Cannabis smoke represents a complex mixture resulting from the pyrolysis of the plant, containing numerous combustion products such as PAHs, nitric oxide, carbon monoxide, and aryl amines12. While Δ9-THC and CBD are the most abundant bioactive compounds found in commercially available cannabis products, studying these cannabinoids in isolation may overlook the complex pharmacological and toxicological properties within whole cannabis smoke26. Indeed, results from this study have identified differences in the effects of THC alone and CaSE on trophoblast cell differentiation and mitochondrial function, which is summarized in Table S2.

We found that Δ9-THC concentrations in our 2.5% and 5.0% CaSE were ~ 1.74 (± 0.18) µM (~ 547.23 ng/mL) and 4.30 (± 0.18) µM (~ 1352.4 ng/mL), respectively, generated from a 12–14% THC-dominant cannabis cigarette (Fig. 1a). Such levels are comparable to those found in cannabis smokers and in human placental tissue (100–432 ng/g), although some of these findings are derived from cannabis cigarettes with lower THC percentages27,28. AhR is primarily activated following exposure to PAHs, such as benzopyrene and benzanthracene, as well aryl-containing amines29, which leads to the induction of its downstream target CYP1A1 – a phase I drug metabolizing enzyme. This enzyme is expressed by placental tissue and is important in metabolizing xenobiotic compounds, including environmental toxicants30. While some studies have demonstrated Δ9-THC itself can inhibit CYP1A1 activity and partially activate its expression through the AhR29, we did not find any significant induction of CYP1A1 expression. Conversely, CaSE treatment resulted in a profound increase in the expression of CYP1A1 transcript levels (Fig. 1c), likely due to the presence of AhR-activating ligands in CaSE. These findings point to the complex mixture effects that result from combustion-derived constituents in cannabis smoke in tandem with the phytocannabinoids. Thus, conclusions drawn from studies using single cannabinoids may overlook critical aspects of xenobiotic metabolism, toxicity and cellular damage associated with cannabis smoke exposure.

Cytotrophoblasts differentiate into a syncytium that plays a critical role in producing and releasing hormones important for initiating, regulating and maintaining pregnancy, an example of which is human chorionic gonadotropin (hCG)31. This hormone consists of two subunits, alpha (CGα) and beta (CGb)32,33. It has been widely recognized, both in vitro and in vivo, that PAH exposure and tobacco smoking are correlated with reduced hCG levels33,34. Despite the conflicting evidence in the literature as to whether cannabis directly impacts serum hCG levels, confirmed pregnant cannabis smokers have shown higher rates of pregnancy loss compared to non-smokers, which may be due, in part, to cannabinoid-induced reductions in hCG levels35,36. Here, we report a dose-dependent reduction of both hCG transcript and protein expression (Fig. 2a) following exposure to CaSE, however treatment with THC alone only caused a significant reduction in CGb expression and not hCG protein. Previous work by our group has shown that despite reduced cell-cell fusion with 10 µM Δ9-THC, only higher THC doses impaired hCG secretion22. Moreover, we observed that transcript levels of ERVW-1 were reduced with both CaSE and 10 µM Δ9-THC treatment. Taken together these data suggest that THC plays a critical role in the differentiation of CTs into the multinucleated syncytiotrophoblast. To understand the differences between THC and CaSE exposure we also measured expression of a proliferative marker, Ki67. CaSE significantly upregulated and THC significantly downregulated the expression of Ki67 (Fig. 2e). During syncytialization, the proliferative capacity of STs is dramatically reduced as differentiation progresses into mitotically inactive cells37. Our results suggest constituents within cannabis smoke may impair differentiation by sustaining a proliferative state, disrupting the transition from CTs into terminally differentiated STs. In fact, with our CaSE treatment, lower concentrations of Δ9-THC (~ 4.30 µM) within the smoke matrix are more damaging to syncytialization compared to single component THC treatment.

Proper syncytium formation requires the coordinated balance of oxidative stress pathways that underpin placental development38. We assessed mitochondrial function to investigate whether cannabis related disruptions to metabolic homeostasis could be contributing to impaired trophoblast differentiation. Previous work from our group has linked impaired mitochondrial respiration and redox signaling to disruptions in trophoblast fusion and hormone production in trophoblast cells39. Mitochondrial dysfunction in the placenta, characterized by oxidative stress and disrupted cellular bioenergetics, has been implicated in poor pregnancy outcomes (e.g. IUGR)40,41, which are consequences that have been associated with prenatal cannabis use42. Oxidative stress occurs from the imbalance of the oxidant and antioxidant systems – this can arise from the excessive production of pro-oxidants and/or the inability of antioxidants to mitigate pro-oxidants43. Cannabis smoke, like cigarette smoke, contains a myriad of pro-oxidants, namely superoxide radicals, nitric oxide and other unstable free radicals44. Additionally, endogenous sources of pro-oxidants, including derivatives of molecular oxygen (O2) like superoxide (O2•−) and hydrogen peroxides (H2O2) produced by the mitochondrial electron transport chain (ETC) can contribute to oxidative stress following smoke exposure45,46. In our study, CaSE significantly increased intracellular ROS levels at low dose CaSE (1% and 2.5%), but ROS levels began to decline at higher doses (5% and 10% CaSE). Mitochondrial membrane potential was preserved at low CaSE doses but decreased at high doses (Fig. 3f). Our dose-dependent changes to ROS and membrane potential may indicate early oxidative stress likely driven by combustion-derived exogenous radicals. Indeed, Sarafian et al. demonstrated in vitro exposure to cannabis smoke, but not Δ9-THC, significantly increased ROS levels, suggesting that pyrolytic byproducts from the smoke may act independently or synergize with cannabinoids to drive ROS production47.

To determine whether the antioxidant defense system may have been primed to respond to this oxidative burden, we evaluated two key antioxidant markers, SOD1 and SOD2, which localize to the cytoplasmic and mitochondrial compartments of the cell, respectively48. Consistent with our observations of differential ROS production at low or high doses of CaSE, we found only SOD1 to be induced with lower CaSE, whereas higher doses of CaSE were associated with upregulations of both SOD1 and SOD2 (Fig. 3c-d). Interestingly, THC alone induced SOD2, indicating a mitochondrial targeted response. Mitochondrial and cytosolic stress markers, HSP60 and HSP70, were differentially regulated (Fig. 3A and B), pointing to the differential intracellular stress pathways that are activated with CaSE versus Δ9-THC. Raja et al. demonstrated the differences to antioxidant stimulation in neuronal SY-SH5Y cells and attributed this to varying THC/CBD ratios present in cannabis extracts49. Importantly, our differences may indeed relate to the inherent antioxidant properties of phytocannabinoids, which are modulated or overlooked in the context of smoke exposure. At higher CaSE concentrations (5% and 10%), decreased ROS production was observed concomitantly with reduced mitochondrial membrane potential, as well as diminished basal respiration rates and ATP production, suggesting ETC dysfunction (Figs. 3f and 4a-b). This is consistent with work showing that both cannabis smoke and Δ9-THC impair mitochondrial function and reduce ATP production in vitro47,50,51. Taken together, these data imply a biphasic metabolic reprogramming and stress response, wherein low dose CaSE induces mitochondrial respiration impairment, accompanied with increasing ROS from ETC derived electron leakage, and corresponding activation of antioxidant defenses. Contrastingly, the higher dose CaSE causes more severe ETC impairment, leading to reduced ROS production resulting from the minimal electron flow and reduced membrane potential. Previous work from our lab has shown cannabinoids (D9-THC and CBD) alter cellular bioenergetics, with the greatest impairments observed at higher D9-THC dosages (10 µM-20 µM) relative to lower D9-THC dosages (1 µM)22,23, which supports the notion that exposure to elevated concentrations of THC induces greater ROS production and attenuation of ATP production, compared to lower doses of D9-THC. Interestingly, this dose-dependent relationship in the present study revealed that lower effective concentrations of D9-THC within CaSE (4.3 µM D9-THC in 5.0% CaSE) elicited equal impairments in mitochondrial respiration compared to single component D9-THC (10 µM) exposure (Fig. 4a-b). These effects were accompanied by alterations in additional Seahorse parameters including increased proton leak and changes in non-mitochondrial respiration, further supporting CaSE-induced disruption of mitochondrial function (Supplemental Fig. S3(a-c)). These results highlight the importance of considering complex smoke mixtures versus isolated cannabinoids when evaluating cellular bioenergetics and trophoblast health. It is also worth noting that impairments in mitochondrial function can impact cellular glucose levels. Although glucose metabolism was not directly explored in this study, future investigations comparing the effects of CaSE and individual cannabinoids on trophoblast glucose metabolism and glycolytic flux could provide valuable insights into these metabolic adaptations. More broadly, our findings that smoke extract-disrupted mitochondrial bioenergetics observed during trophoblast differentiation may have important implications for proper placental development and efficiency39.

Cannabinoid receptor signaling is instrumental during pregnancy and therefore warrants careful consideration in the context of prenatal cannabinoid exposures52. For the purposes of this study, we focused on CB1 receptor pharmacology as a proxy to reveal the complex ECS receptor changes induced by Δ9-THC and smoke exposure. CB1 is one of the most well-known receptors in the ECS and has a widely recognized role in trophoblast differentiation and overall reproductive success52,53,54,55. Δ9-THC and its corresponding metabolite, 11-hydroxy-THC, are partial agonists of the CB1 receptor, making this a relevant target in the context of cannabis extract exposure56. We report that protein levels of CB1R do not change following treatment with either Δ9-THC treatment or smoke extract (Fig. 5a). Chronic cannabis use has been shown to desensitize the CB1 receptor, without reducing receptor protein levels, which is consistent with our findings57. Our observation that 10 µM Δ9-THC does not alter CB1 receptor protein expression is also consistent with reports following both 24 h and 72 h exposure in placental explants58. In our study, incubation with a CB1 receptor antagonist revealed that Δ9-THC impacted trophoblast differentiation through CB1 receptor-mediated signaling (Fig. 5d-f) an effect which is consistent with previous work from our group22. In the presence of the CB1-receptor antagonist AM281 with CaSE, there was neither blockage nor attenuation of any adverse effects related to reduced differentiation markers induced by the CaSE treatment (Fig. 5d-f). Therefore, the exposure to AhR ligands generated by the CaSE treatment may point to potential pathways that smoke may be acting through to cause the adverse effects related to differentiation59.

While this study provides novel insights into cannabis smoke-induced dysfunction in trophoblasts, there are some limitations to consider. BeWo b30 cells are a third-trimester choriocarcinoma cell line that resembles cytotrophoblast-like cells capable of differentiating into syncytiotrophoblast, which are cell populations that reside at the maternal-fetal interface. The differentiation of cytotrophoblasts into syncytiotrophoblast is a process that occurs continuously throughout gestation, underscoring the relevance of this model in capturing the dynamic nature of placental biology60,61. We modeled this differentiation process over a 48-hour period, as this timeframe is commonly used to capture syncytialization in vitro61. This approach allowed us to assess the direct cellular effects of combustion-derived constituents and cannabinoids on trophoblast physiology during differentiation in a controlled environment. Our 48-hour treatment window represents an acute exposure at the cellular level rather than a chronic in vitro exposure, providing insight into the immediate effects of cannabis smoke extract on trophoblast function. Future studies could expand upon these findings by investigating both short- and long-term exposures to model acute versus chronic cannabis smokeexposure. Additionally, examining the baseline effects of CaSE and D9-THC on undifferentiated cytotrophoblast cells prior to syncytialization would help delineate initial cellular changes that precede the differentiation-associated responses.

In the present study, we showed cannabis smoke extract differentially alters key pathways involved in trophoblast metabolism and differentiation. Our work offers insights into the complexity of cannabis smoke versus treatments with synthetic cannabinoids such as Δ9-THC, and underscores the importance of including combustion-derived constituents in reproductive smoking models. Taken together, our findings demonstrate the urgent need to evaluate cannabis smoke as a toxicological exposure route in reproductive health. It also provides valuable insights for understanding the mechanistic underpinnings of cannabis smoke-induced dysfunction in trophoblast cells and develop harm reduction strategies for prenatal cannabis use. Future studies should consider cannabinoid profiles that mirror the evolving cannabis market as well as routes of exposure such as smoking, which persists as the primary route of self-administering cannabis during pregnancy1.

Methods

Cell culture

BeWo b30 cells (AddexBio, C0030002) were cultured in in 1X Dulbecco’s Modified Eagle Medium (1X DMEM) containing 4.5 g/L glucose, L-glutamine, sodium pyruvate supplemented with 10% heat-inactivated fetal bovine serum (HI-FBS), 1% penicillin-streptomycin and 1% L-glutamine, incubated at 37 °C and 5% CO2 conditions. Cells were plated on multi-well plates at a seeding density of approximately 1 × 105 cells/cm2 to achieve ∼90% confluency. Following 24 h of cell adherence, epidermal growth factor (EGF; 50 ng/mL) was added to the media to stimulate monolayer formation and proliferation. After 48 h, cells were treated with forskolin-1 (FSK; 50 µM) along with additional EGF to induce trophoblast differentiation and fusion of syncytiotrophoblast (ST) for a 48-hour duration. At the time of FSK and EGF addition, cells were treated with either vehicle control (0.1% methanol), 1 µM or 10 µM Δ9-THC, or 1.0%, 2.5%, 5.0% or 10% cannabis smoke extract (CaSE). Smoke-treated cultured cells were kept in a separate incubator at 37 °C and 5% CO2 conditions for the duration of the exposure.

Preparation of cannabis smoke extract

THC-dominant, dried cannabis flower was purchased from Cannalogue Labs (14% THC: 1% CBD). Cigarettes were hand-packed into king size cigarette tubes, each containing ∼0.70 g of dried cannabis. Prior to use, the filter of each cigarette was removed. Briefly, CaSE was prepared by drawing smoke, under vacuum, through 10 mL of cell culture media. On average, the burn duration for each cigarette was 2:30 min. The CaSE was then passed through a 0.22 μm filter and diluted either 1:100 (1%), 1:40 (2.5%), 1:20 (5%), or 1:10 (10%), using fresh cell culture media (described above). Fixed combustion times, burn rates, and day-of-use were parameters consistently and carefully considered to ensure reproducibility.

Detection and Quantitation of Cannabinoids by LC-MS/MS

Smoke extract media (before and after filtration) and sample supernatants were analyzed using targeted liquid chromatography–tandem mass spectrometry (LC–MS/MS) for the quantification of phytocannabinoids (THC and CBD). Native and isotopically labeled standards (d₃-THC and d₃-CBD; purity > 98%) were obtained from Toronto Research Chemicals and stored at 4 °C.

Method performance was evaluated by spiking in vitro media (60 µL) with THC and CBD at three concentrations (10, 500, and 1500 pg/µL; n = 6 for each level), followed by dilution to 1500 µL with acetonitrile. Across all spike levels, analyte recoveries exceeded 90% and repeatability was better than 15%. Limits of detection (LODs) were determined in accordance with the Eurachem Guide to Method Validation using the equation:

$$\:\text{LOD}=3\times\:{s}_{o}^{{\prime\:}}\times\:\sqrt{1/n}$$

where \(\:{s}_{o}^{{\prime\:}}\)is the adjusted standard deviation calculated from replicate measurements of media spiked at 10 pg/µL63. Resulting LODs were 7.5 pg/µL for THC and 7.1 pg/µL for CBD. Limits of quantitation (LOQs) were determined by substituting 10 for the factor of 3 in the equation above: yielding LOQs of 25.3 pg/ µL for THC and 23.5 pg/µL for CBD. For sample quantification, 60 µL of in vitro media was transferred to an LC vial, fortified with 20 µL of d₃-THC and d₃-CBD internal standards (7.5 ng/µL), and diluted to a final volume of 1500 µL with acetonitrile. Chromatographic separation was performed on an XSelect HSS T3 C18 column (2.5 μm, 2.1 mm × 50 mm) using a binary mobile phase consisting of methanol and water. Mass spectrometric detection was carried out using positive electrospray ionization on an AB Sciex 365 triple quadrupole instrument equipped with an HSID Ionics Eclipse 10 Plus orthogonal ionization source. Additional details, including the HPLC gradient and multiple reaction monitoring ion transitions used for THC and CBD quantification, are provided in Brown et al. (2019)58.

Cell viability

Cells were seeded and treated in 96-well plates with varying dilutions of CaSE as described above. Following treatment, 10% AlamarBlue (Bio-Rad Laboratories) was added to fresh media, and cells were incubated at 37 °C in 5% CO2 for 4 h, protected from light. Absorbance was read at 570 nm and 600 nm for test and blank wells, respectively, in a microplate reader (Tecan Spark® Multimode Microplate Reader).

RNA extraction & real-time quantitative PCR

Total RNA was isolated from BeWo b30 cells grown, treated, and harvested from 6-well plates as previously described. Briefly, cells were lysed with 500 µl of ice-cold TRIzol ™ reagent (Thermo Fisher Scientific) and total RNA, was extracted using the Direct-zol RNA MiniPrep Kit (Zymo Research) according to the manufacturers protocol. ~1 µg of RNA was converted to complementary DNA (cDNA) using the High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems™) as per the manufacturer instructions. Quantitative real-time PCR was used to measure mRNA expression detected with GB-Amp™ SYBR Green qPCR Mix (GeneBio Systems) using the CFX384 Touch TM Real-Time PCR Detection System (Bio-Rad Laboratories) Analysis of fold-change in mRNA expression levels was calculated using the double-delta (2−ΔΔCT) method, normalized to the geometric means of reference gene expression levels (18 S, β-Actin) and then to the vehicle control. Primer sequences can be found in Supplementary Table S1.

BCA (Bicinchoninic Assay)

Protein concentration of samples was determined using the bicinchoninic assay (BCA; ThermoFisher). Total protein concentration was measured at an absorbance of 562 nm in a 96-well plate using a microplate reader (Tecan Spark® Multimode Microplate Reader).

SDS-PAGE and Western blotting

Cells treated with Δ9-THC and CaSE were harvested using radioimmunoprecipitation assay (RIPA) lysis buffer, containing phosphatase and protease inhibitors (Roche Diagnostics). Samples were prepared from total protein extract, Laemelli Sample Buffer (2X) and β-mercaptoethanol. Proteins (20–25 µg/sample) were separated by gel electrophoresis using Mini-PROTEAN® TGX Stain-Free™ gels. Following electrophoretic separation, the gels (Bio-Rad Laboratories) were transferred onto PVDF membranes using the Trans-Blot® Turbo™ Transfer System (Bio-Rad Laboratories). Membranes were blocked in 1X Tris-buffered saline-Tween 20 buffer with 5% BSA and probed using anti-hCG (DAKO; Cat. #: GA508; Lot #: 20016637; rabbit polyclonal; 1:1000) and anti-CB1R (Cayman; Cat. #: 10006590; Lot #: 0577427-1; rabbit polyclonal, 1:5000) in the blocking solution. Donkey anti-rabbit (1:10,000) secondary antibodies were used for immunodetection of the primary antibody. Proteins were visualized using Clarity Max Western ECL Substrate (Bio-Rad Laboratories) and visualized using ChemiDoc Imaging System (Bio-Rad Laboratories). Band intensities were quantified using Image Lab (Bio-Rad Laboratories). Proteins of interest were normalized to total protein content on the corresponding membrane following visualization with stain-free imaging technology.

DCFDA assay (2′,7′-dichlorofloroescin diacetate)

Reactive oxygen species (ROS) were detected using the DCFDA/ H2DCFDA Cellular ROS Detection Assay Kit (Abcam, ab113851) as per the manufacturer’s instructions. Briefly, cells were seeded in a black, clear bottom 96-well microplate at a cell density of 32,000 cells; differentiated and treated as described above. Following 48 h of treatment, media was removed, and the cells were washed with 1X Buffer (90% ddH2O, 10% 10X Buffer), then incubated with DCFDA (5 µM) for 45 min. The resulting fluorescent signal wasquantified at an excitation/emission of 485/535nm (Tecan Spark® Multimode Microplate Reader). All readings were normalized to total protein content via the BCA assay.

Mitochondrial respiration assay

The mitochondrial oxygen consumption rate (OCR) was measured at 37 °C in an XFe24 Extracellular Flux Analyzer (Agilent Bioscience). BeWo b30 cells were seeded at 50,000 cells/well in 250 µl culture media in 24-well microplates; and were differentiated and treated as described above. Following 48 h, cultured media was removed and replaced with XF base medium (Seahorse Bioscience) supplemented with 100 mM sodium pyruvate, 200 mM L-glutamine, and 5 mL of 45% glucose solution, warmed to 37 °C (pH 7.4). The assay medium was then pre-equilibrated to 37 °C for 1 h. The OCR was detected under basal conditions followed by the sequential injection of oligomycin (ATP synthase inhibitor), carbonyl cyanide-4- (trifluoromethoxy)phenylhydrazone (FCCP; mitochondrial uncoupler) and rotenone combined with antimycin A (complex I and III electron transport blockers) through ports of the Seahorse Flux Pak cartridges to achieve final concentrations of 1, 0.3 and 0.5 µM, respectively. OCR values were normalized to total protein content from each well determined through the BCA assay. Each OCR detection experiment was performed with four technical replicates per experiment, for each treatment condition, to obtain an average response per experiment. Each experiment was repeated 3 times.

Mitochondrial membrane potential

Mitochondrial membrane potential (ΔΨm) was assessed using the fluorescent reagent tetraethylbenzimidazolylcarbocyanine iodide (JC-1) with the JC‐1‐Mitochondrial Membrane Potential Assay kit (Abcam, ab113850) following the manufacturer’s instructions. Briefly, cells were seeded in a black, clear bottom 96-well microplates at a cell density of 32,000 cells, differentiated and treated as described above. Following 48 h, cells were washed once with 1X dilution buffer (90% ddH2O, 10% 10X Dilution Buffer) and incubated with the JC‐1 dye (20 µM) in 1X dilution buffer for 10 min at 37 °C, protected from light. The dye was then removed, cells were washed once with 1X dilution buffer and 100 µl of 1X Supplemented Buffer (2% FBS, 98% ddH2O) was added. The red fluorescence intensity (excitation/emission (535 nm/590 nm)) and green fluorescence (excitation/emission (475 nm/530 nm)) were measured using a microplate reader (Tecan Spark® Multimode Microplate Reader). Background fluorescence was subtracted, and the ratio of red (polarized) fluorescence divided by green (depolarized) fluorescence was calculated.

Evaluation of antagonism at CB1 receptor(s)

The efficacy of Δ9-THC and CaSE at CB1 receptors on BeWo b30 cells was evaluated by pre-incubating with a selective CB1 antagonist, AM281 (1 µM) for 30 min prior to drug treatments. The concentration and incubation period were selected based on previous studies demonstrating effective CB1R blocking at 1 µM22,64,65. Following the pre-incubation, a subset of cells was re-incubated with the AM281 and corresponding treatment; cells were treated and differentiated as described above.

Statistical analyses

All statistical analyses were performed using GraphPad Prism software V6.0. All experiments were performed with six biological replicates (unless otherwise specified). All outliers were identified with a Grubbs’ test (alpha = 0.05) followed by an assessment for normal distribution using the Shapiro-Wilk test. A one-way ANOVA (analysis of variance) was used to compare differences among two or more groups, followed by a post-hoc Tukey test for multiple comparisons. Data is reported as means ± S.E.M., unless otherwise stated. Different letters denote significant differences, which is defined as a statistical threshold of p < 0.05.