In mammary cancer, high-grade tumors are often treated with chemotherapy as an adjuvant and palliative strategy. However, this therapeutic alternative is not selective and can induce several side/adverse effects. CBD, a compound with demonstrated antineoplastic capacity against various malignant tumors [9,26], may represent an alternative. However, applications of CBD are limited due to low solubility in biological media, instability, and poor cellular uptake and internalization [35,36]. These CBD characteristics also contribute to the variability in experimental results reported in the literature [37]. In this study, we prepared a nanoemulsion previously formulated by our team [24], which was able to solubilize CBD (in the oil nucleus of the nanoemulsion), allowing the dispersibility of CBD in biological media, protecting the molecule from environmental adversities, and increasing the cellular uptake and internalization. The oil used (Miglyol 812, Sasol GmbH-Germany, Hamburg, Germany) corresponds to a medium-chain triglyceride (MCT) where CBD shows high solubility (273.8 mg/mL), which allowed a complete dissolution of the drug [17]. In addition, CBD-Nem was developed avoiding the use of toxic solvents and employing FDA-approved excipients, making it suitable for in vivo applications in preclinical and clinical protocols. Unlike other strategies used to create CBD-Nem, such as those that employ sonication and two-step high-pressure homogenization [38], the methodology described here (solvent displacement) is less polluting, consumes less energy, and is more easily scalable.

CBD-Nem had a spherical shape (Figure 1A), size in the range of ~150 nm, with polydispersity ≤ 0.15, and zeta potential of ~−50 mV (Table 1). These values are ideal since they contribute to the stability of the nanosuspension in biological medium, preventing self-aggregation [39]. These results resemble those found in other CBD-loaded nanoemulsions, as described by Lewinska (2021) [38] and Fathordoobady et al., 2021 [40], where zeta potential values around −50 mV were achieved, using surfactin obtained from Bacillus subtilis natto (KB1) and lecithin as surfactants. Unlike our formulation, the referred studies utilized high-energy techniques (sonication, high-pressure homogenization, ultrasonication, and microfluidization) to produce CBD-loaded nanoemulsions. In addition, as evidenced in Figure 1B, the proposed nanoformulation in this paper is highly stable in aqueous medium. Furthermore, the concentration of nanocarriers (nanoparticles per mL) obtained in this work is in the superior range (~1–5 × 10¹³) as presented in other works developing nanoparticles [27,28,41,42], representing an efficient methodology for industrial purposes.

CBD-Nem exhibited very high encapsulation efficiency (99%, Table 1), which is equivalent to 5 mg/mL. Other formulations, such as poly (lactic-co-glycolic acid) (PLGA) nanocapsules loaded with CBD, have reached encapsulation efficiencies around 95% and encapsulating between 0.1 to 0.8 mg CBD/10 mg PLGA [23].

The release kinetics of CBD from Nem in PBS were tested at pH 5.5 and 7.4, with no detectable concentrations until the third and sixth day, respectively (Figure 2). However, this lack of detection could be related to the detection limit of CBD obtained through UV spectroscopy, which was 0.05 mg/mL. In this regard, the use of chromatographic techniques with greater analytical sensitivity could improve detection thresholds [43]. The release assay described here showed a slow release, slightly faster at pH 5.5, but without significant differences until day 34. The release model with the highest fit, according to the observed determination coefficient (R² = 0.9), was the Hopfenberg kinetics model, indicating that the main limiting process in the release of CBD from nanoemulsions was matrix erosion [44]. The slow release of CBD from nanoemulsions could be related to the high lipophilicity of the CBD molecule, which results in a greater affinity for the oily core than for the external medium [45]. Additionally, particle size may also play a role, as described by Aparicio-Blanco et al., 2018 [20], who reported slower CBD release from 50 nm PLGA nanocapsules compared to 20 nm ones. PLGA microcapsules of 24 µm have also shown sustained release, with 80% of CBD released after 42 days [18]. Other studies using CBD in Nem and manufactured using microfluidic and ultrasound energies (using lecithin as emulsifier and poloxamer 188 as co-emulsifier) yielded significantly faster drug release profiles. Using simulated gastric fluid (SGF, pH = 1.3) and simulated intestinal fluid (SIF, pH = 7.0), abrupt release rates were observed within the first hour of exposure at both pH levels, followed by a reduced release rate during the subsequent 3 and 12 h, respectively, with a more pronounced effect in SGF [40]. Duse et al., 2019 [46] reported that the effect of pH on lipophilic drug release depends on the erosion contribution as the release mechanism, describing faster release from PLGA nanoparticles at pH 5.5 compared to 7.4.

Rho was used to indirectly assess CBD internalization in CF41.Mg cells due to physicochemical similarities with CBD (Rho: molecular weight of 479 g/mol and a LogP of 6.3; CBD: molecular weight of 315 and LogP: 6.3). The results of FACS showed a more efficient uptake when encapsulated in Nem compared to the ethanol vehicle (Figure 3A). Rho-Nem achieved 29% higher fluorescence in CF41.Mg cells than Rho-E. These results were further supported by a confocal microscopy evaluation, showing higher intracellular fluorescence in CF41.Mg cells treated with Rho-Nem compared to those treated with Rho-E (Figure 3B). This finding agrees with a previous study using the lipophilic and fluorescent curcumin loaded in Nem, which reported a significant increase in uptake (22%) and internalization (evidenced by a stronger intracellular fluorescent signal of curcumin) than curcumin dissolved in DMSO, demonstrating enhanced endosomal uptake [24]. However, this study represents a preliminary and indirect evaluation of CBD uptake and internalization. Since Rho is a charged molecule (like curcumin), unlike CBD, and is not structurally similar to CBD, the mechanisms of uptake and internalization may differ. Facilitated transport or endocytosis may play a more prominent role in Rho internalization, whereas CBD could also enter cells through passive diffusion [47]. Previous studies have shown that a negative surface charge on nanocarriers (−30 to −50 mV) does not hinder cellular internalization [24,27,48,49,50]. In addition, the nanoscale size of these carriers (100–250 nm) facilitates internalization [24,51,52]. Furthermore, the intrinsic permeation properties of the phospholipids used for stabilization in this paper (surfactant of soybean lecithin) also favor the internalization of Nem [27].

To date, nanoemulsions containing CBD have not been evaluated on canine mammary cancer cells. In this work, the canine cell line CF41.Mg and IPC366 were selected. In addition, the human mammary cancer cell line MDA-MB-231 was also tested to compare the results and show the potential for this formulation in human applications. The cell line MDCK (canine kidney cells) was used as a control of non-cancer cells. Cell viability was assessed by seeding all cell lines at the same density (1.5 × 10³) to ensure reproducible and comparable results. We observed that CBD-Nem reduced the viability of mammary carcinoma cells, effects that were time- and concentration-dependent. The reduced response of CF41.Mg cells to CBD-Nem (Figure 4A) compared to the ethanol vehicle (Figure 4B). Additionally, CF41.Mg cells exhibited lower susceptibility to the effects of CBD-Nem compared to IPC366 and MDA-MB-231 cells, where a greater decrease in viability was observed at 20 µM (Figure 5A,B), although these effects were less pronounced than those of CBD-E in both cell lines. Nevertheless, it is relevant that the cytotoxic effect of CBD-Nem is reproducible in different types of canine and human mammary tumor cells representative of high-grade tumors. The less pronounced effect of CBD-Nem compared to CBD-E on the cell viability of the evaluated cancer cell lines, without achieving an IC₅₀ even at the highest tested dose (50 µM), may be attributed to several factors. One of them is the slow release of CBD from the nanoemulsions (Figure 2), which could hinder, within the evaluated time frame, the interaction of CBD with membrane receptors such as GPR55 and TRPVs, which can modulate the viability of cancer cells. Moreover, the nanoemulsion may be fully internalized by cancer cells, as nanocarriers smaller than 500 nm—both cationic and anionic—are internalized via pinocytosis mechanisms [47]. This could delay the interaction of CBD with intracellular receptors until it is released from the nanoemulsion. The nanoemulsion would enhance CBD internalization into cancer cells, as suggested by outcomes with Rho-Nem. In this regard, we hypothesize that increased internalization may enhance interactions of CBD with intracellular targets such as PPARγ (involved in the induction of apoptosis), mitochondrial receptors like the voltage-dependent anion channel 1 (VDAC1), and endoplasmic reticulum calcium channels [9]. In contrast, ethanol as a vehicle may favor CBD interactions with membrane receptors such as TRPV, GPR55, and CB1-2 [53]. For this reason, the nanoemulsion could show better results at longer time points, as observed in the colony formation assays.

Interestingly, CBD-Nem triggers different effects on the viability of CF41.Mg (Figure 4A) compared to MDCK cells (Figure 6). Cytotoxic effects were observed in carcinoma cells over time, while non-tumor cells showed recovery of viability at 72 h (Figure 6C), indicating a protective effect (related to safety) for the CBD-Nem formulation. This differential effect could be explained by differences in the CB1-CB2 receptor expression in both cell lines. MDCK cells express higher levels of CB1 than CB2 receptors [54], and CF41.Mg cells have higher CB2 expression than CB1. In human breast cancer, CBD has been reported to inhibit cell growth and induce apoptosis through the activation of both CB2 and TRPV1 [55]. CB1 activation in renal tissue is associated with Na+/K+ ATPase pump activation and mitochondrial fission stimulation [56,57]. In addition, with the aim of mimicking a medicine with prolonged activity, a longer antineoplastic effect (20 days) was tested by administering CBD-Nem 20 and 50 µM. The formulation demonstrated inhibition in the colony formation ability of CF41.Mg cells (Figure 7). To the best of our knowledge, there are no other studies using Nem containing CBD, or other CBD formulations comprising only FDA-approved components and demonstrating efficacy in clonogenic studies.

In terms of the cellular phase, it was also observed that CF41.Mg cells treated with CBD (whether formulated in ethanol or Nem) remained in a higher proportion in the G0–G1 phase, with no significant differences compared to the control (Figure 8A). Additionally, a reduction in the proportion of CF41.Mg cells in the G2-M phase were noted, which suggests a cell cycle arrest at the S phase. Concordantly, an increase in the S phase was observed in response to 50 µM CBD-Nem, which indicates an arrest of DNA replication. In this context, the intracellular DNA repair mechanism is activated, triggering phosphorylation of the CDK2/Cyclin E complex, which leads to prolongation of the S phase and would provide more time for DNA repair [58]. Previous studies in human breast cancer cell lines (MCF-7, ZR-75-1, T47D, MDA-MB-231, MDA-MB-468, and SK-BR3) have shown that treatment with CBD delivered in DMSO induces G0-G1 arrest through CB1 receptor activation, leading to downregulation of cyclin D1 [55,59,60]. Additionally, activation of the CB2 receptor has been reported to affect cell cycle progression by inhibiting Cdc2, a key regulator of cell division, thereby inducing cell cycle arrest in the G2/M phase [60,61,62]. Another study demonstrated that CBD arrests the cell cycle of MDA-MB-231 cells in the late S-G2 phase rather than in G1 [63]. Consistent with these findings, CBD has been shown to exert antiproliferative activity by inducing cell cycle arrest at different phases depending on the concentration used, suggesting the involvement of distinct mechanisms of action underlying its antiproliferative effects [60,63]. In the present work, we demonstrate that this biological potential for CBD during cellular phases is maintained, although the drug is formulated in the form of CBD-Nem.

Moreover, a significant increase in apoptosis in CF41.Mg cells treated with 20 and 50 µM CBD-Nem for 24 h were evidenced; this increase is similar to that observed when using the positive control doxorubicin (1 µM) (Figure 8B). Interestingly, these results contrast with those obtained at 24 h in the MTS reduction assay (Figure 4A), which showed a non-significant 5% decrease in CF41.Mg cell viability following treatment with CBD-Nem at 20 µM. Considering that both assays evaluate distinct cellular processes (MTS is associated with cellular respiration, while the apoptosis assay detects extracellular exposure of phosphatidylserine on the cell membrane), the results may reflect biological overlapping processes at the 24 h time point. Importantly, the initiation of apoptosis can be reversed if the cell manages its homeostatic balance [64]. In another study conducted by Henry et al., 2020 [10], the canine mammary carcinoma cell line CMT12 treated with 11 µM of CBD (in ethanol) showed a significant increase in annexin V-positive cells after 8 h. Furthermore, CBD delivered in DMSO has been reported to induce an apoptotic effect involving caspase-3 in MDA-MB-231 cells [55]. Additional evidence indicates that CBD exerts pro-apoptotic effects by downregulating mTOR, AKT, 4EBP1, and cyclin D, while upregulating the expression and nuclear translocation of PPAR [13,59].

As shown in Figure 9, the exposure to CBD-Nem 20 and 50 µM also prevents migration and invasion in CF41.Mg cells. Similar effects have been observed in human cell lines 4T1.2 and SUM159 treated with 6 µM CBD dissolved in DMSO [8]. McAllister et al. [65] described that CBD negatively regulates the production of ROS and the Id-1 protein in MDA-MB-231 cells, producing a decrease in migration and invasion. It was also observed that metalloproteinases, phalloidin, and actin stress fibers, which are important in tumor invasion, were suppressed by CBD [65]. However, the specific CBD receptors involved in these effects are unknown.

Although this study shows that our CBD-Nem has antitumor effects on canine mammary cancer cells, further research is needed to elucidate its therapeutic potential. The uptake/internalization study could be improved by using antibodies against cannabidiol to directly label it, as well as antibodies for co-localization of endosomal pathways, such as anti-EEA1 and anti-Rab5, among others. These studies could be complemented with mechanistic assays to investigate the pathways underlying the effects of CBD-Nem. On the other hand, in vivo pharmacokinetic studies of CBD-Nem in tissues and animal models should be conducted to evaluate the concentrations reached and the persistence of the formulation in tissues. All this information supports the performance of clinical trials in dogs with mammary carcinomas to define the safety and efficacy of this formulation.