To the best of our knowledge, this is the first study that has provided functional evidence about the adenosine transport inhibitory effect of CBD, a promising non-intoxicating phytocannabinoid, in the myocardium. The chronic oral administration of CBD to obese ZDF rats significantly increased the response of their isolated, paced left atria to adenosine, suggesting the presence of a long-term enhanced adenosinergic protection in the heart.

CBD has been being extensively investigated for a variety of indications owing to its antiinflammatory, antitumor, neuroprotective, anticonvulsant, anxiolytic, antipsychotic, antidepressant and antidiabetic properties (Bielawiec et al., 2020; Sunda and Arowolo, 2020; Kicman and Toczek, 2020; Bilbao and Spanagel, 2022; Peng et al., 2022). Moreover, CBD is commercially available as a dietary supplement. Thus, it is crucial to explore the consequences of the long-term use of CBD, with special regard to vital but vulnerable organs, such as the heart, and to fundamental protective mechanisms, like the adenosinergic system. Thus, in the present study, we focused on the effect that long-term oral CBD treatment exerted on the myocardial adenosinergic function, which was quantified by the robust direct negative inotropy characteristic of the atrium (Belardinelli et al., 1995; Fredholm et al., 2001). In the myocardium, most effects of adenosine, including the direct negative inotropy, are mediated by the A₁ adenosine receptor (Belardinelli et al., 1995; Fredholm et al., 2001; Szentmiklosi et al., 2011; Headrick et al., 2013; Lasley, 2018).

Our CBD administration protocol aimed to model the human use of CBD, with special regard to the antidiabetic indication. In most human studies, the repetitive, oral administration of CBD was carried out with doses between about 1 and 50 mg/kg/day for 1–6 weeks(s) (Millar et al., 2019; Kicman and Toczek, 2020). Using the formula “rat dose = human dose ∙ 6.2” (Nair and Jacob, 2016), a rat dose range 6.2–310 mg/kg/day was obtained. In mice and rats, the most used repetitive administration regimens consisted of 10 mg/kg/day CBD, i. p. Or per os, for 1–6 weeks(s) (Wheal et al., 2017; Karimian Azari et al., 2020). Interestingly, i. p. And oral routes were found to result in similar concentrations in the rat (both in the plasma and brain) (Deiana et al., 2012). In diabetic rodent models, the duration of CBD treatment varied between 1 and 11 weeks(s), but the most frequent length was 4 weeks (El-Remessy et al., 2006; Weiss et al., 2006, 2008; Rajesh et al., 2010; Santiago et al., 2019). Taking all together, we decided to use 60 mg/kg/day CBD, per os, for 4 weeks, considering both the dose and duration to be sufficient to act in a well-detectable but safe (Dziwenka et al., 2020) manner.

In the present study, CPA, a relatively stable (Pavan and IJzerman, 1998) and thereby little transported, synthetic A₁ adenosine receptor agonist (Fredholm et al., 2001, 2011), evoked a significantly smaller response in the atria isolated from the CBD-treated obese ZDF rats than in the atria of the vehicle-treated ones (Figure 1, left panel). In contrast, adenosine, the rapidly metabolized and transported physiological adenosine receptor agonist (Pavan and IJzerman, 1998; Fredholm et al., 2001, 2011), elicited a significantly greater response in the CBD-treated atria than in the vehicle-treated ones, but only at low and medium concentrations (Figure 1, right panel). According to the earlier experiences of our work team, this pattern is typical of the condition when the myocardial adenosine transport is blocked (Karsai et al., 2006; Kiss et al., 2013; Erdei et al., 2018; Viczjan et al., 2021). The reason for this is as follows: Upon sufficient oxygen supply, the myocardial adenosine transport is directed into the cells, so its inhibition increases the interstitial level of endogenous adenosine (Deussen et al., 1999, 2006; Karsai et al., 2006). This surplus interstitial adenosine, in part, uses up the response capacity of the adenosine receptors (prior to the generation of an E/c curve) (Gesztelyi et al., 2004). Hence, a poorly transported adenosine receptor agonist (like CPA) cannot evoke an effect as great as that seen with intact transport. Contrary to this, when a quickly metabolized and transported adenosine receptor agonist (like adenosine) is administered, two opposing effects prevail: the above-mentioned distorting effect that virtually reduces the response, and another effect that really augments the response, since the transport blockade protects also this metabolizable and transportable exogenous agonist from the intense intracellular elimination. As a result of these two opposing effects, enhanced responses to low and medium agonist concentrations, furthermore reduced or unchanged responses to high agonist concentrations develop. This was observed in our previous studies performed in guinea pig and rat atria, using nitrobenzylthioinosine derivatives for adenosine transport blockade and applying adenosine as a metabolizable and transportable exogenous agonist (Karsai et al., 2006; Kiss et al., 2013; Erdei et al., 2018; Viczjan et al., 2021). This is also the case in the present study as regards the CBD-treated atria: we have found weaker responses to all CPA concentrations (Figure 1, left panel), while stronger responses to low and medium adenosine concentrations and unchanged responses to high adenosine concentrations (in comparison with the vehicle-treated atria) (Figure 1, right panel).

Indeed, CBD was found to inhibit ENT1 (Carrier et al., 2006; Liou et al., 2008; Pandolfo et al., 2011). This inhibitory effect was reported to be relatively strong and concentration-dependent, to evolve from relatively low, 100 nM concentrations (Carrier et al., 2006; Ibeas Bih et al., 2015), and it was evidenced in neurons (Pandolfo et al., 2011), macrophages, retinal and brain microglial cells (Carrier et al., 2006; Liou et al., 2008), but not in the heart. Our present study is the first to provide functional evidence about the ENT1 inhibitory effect of CBD in the myocardium.

As adenosine activates all types of adenosine receptors, the possible involvement of the cardiac A_2A adenosine receptor should also be addressed. The A_2A adenosine receptor, primarily by stimulating the adenylyl cyclase, may increase the contractile force (Boknik et al., 2021). If CBD were able to attenuate the function of A_2A adenosine receptor (or its signaling), this mechanism could contribute to the stronger negative inotropic effect of adenosine after CBD treatment that was observed in the present study (Figure 1, right panel). However, to the best of our knowledge, no finding has yet been described to suggest this. Similarly, apart from suggestions (e.g. see Table 1 in Kicman and Toczek, 2020), no evidence about effects of CBD on activities of adenosine-handling enzymes has been found so far. Furthermore, if we deliberate the conceivable effects of CBD on the myocardial adenosinergic system other than nucleoside transport, none of them can explain our finding, i.e. CBD exerted opposite effects on the responses to adenosine and CPA (Figure 1).

To properly exhibit and evaluate the ENT1 inhibitory effect of CBD, some adjustment on the E/c curves representing the CBD-treated condition was needed. As mentioned above, if neglected, the presence of a surplus interstitial adenosine biases the conventionally evaluated and plotted E/c curves of adenosine receptor agonists, namely it causes a virtual (and not real) decrease in the response to these agonists (that can be spectacular or barely noticeable, depending on the circumstances). This distortion can lead to misinterpretations when comparing the affected E/c curves with undistorted (inherently correct) ones. To solve this problem, previously we elaborated a method, by which this E/c curve distortion can be corrected (Kiss et al., 2013; Erdei et al., 2018). With this method, it is possible to obtain E/c curve effect values that provide information on the magnitude of the cause of the distortion (looking at the initial part of the corrected E/c curve) and on the real receptorial responsiveness (looking at the final part of the corrected E/c curve).

The corrected CBD-treated CPA E/c curve started from an about 25% effect value, indicating that CBD accumulated circa the EC₂₅ value of adenosine in the microenvironment of the myocardial A₁ adenosine receptors (in terms of the direct negative inotropic effect) (Figure 3, left panel). Furthermore, the final part of the corrected CBD-treated adenosine E/c curve shows a considerably stronger maximal response than that of the vehicle-treated counterpart (Figure 3, right panel). These results imply that the long-term oral CBD treatment maintains a continuous, moderately elevated basal adenosinergic activity in the heart, furthermore it enhances the response to exogenous adenosine as well as to stimuli leading to adenosine release in the myocardial interstitium.

Nevertheless, it should be noted that our correction method ascribes all distortions manifested in a smaller response to an extra, unaccounted agonist concentration, which is here the surplus interstitial adenosine concentration accumulated by CBD (Figure 2). However, because of the chronic presence of a surplus adenosine, the issue of downregulation and/or desensitization of the A₁ adenosine receptor should also be addressed. In a previous work, we found that the responsiveness of the myocardial A₁ adenosine receptor did not decrease in the time window of our ex vivo experiments (Gesztelyi et al., 2004). Based on this, in our earlier studies dealing with the acute consequences of adenosine transport inhibition (Karsai et al., 2006; Kiss et al., 2013; Erdei et al., 2018; Viczjan et al., 2021), no decrease in the (real) responsiveness of the A₁ adenosine receptor was considered. Consistent with this, the A₁ adenosine receptor was reported to be desensitized extremely slowly, even in the presence of significant amounts of full agonists for several weeks (Willems et al., 2006; Mundell and Kelly, 2011). The 4-week CBD treatment applied in the present study, however, was long enough to consider the possibility of the desensitization of the A₁ adenosine receptor. If there was some desensitization, the response to adenosine after the four-week CBD treatment may have been not as enhanced as it looks on the corrected adenosine E/c curve (Figure 3, right panel). The reason for this is that, upon desensitized receptors, the decrease in the response stems in part from the receptor desensitization, so the correction method overestimates the concentration of the surplus agonist that leads to an overcorrection of the distorted E/c curve. However, taking a look at the conventionally evaluated adenosine E/c curves (Figure 1, right panel), the long-term CBD treatment indisputably increased the response to adenosine at concentrations from 1 nmol/L to 10 μmol/L, an important range regarding cardioprotection (Headrick, 1996; Lasley and Mentzer, 1998) (So, what may be disputed here is only that the decreasing effect, which affected the even so enhanced response to adenosine, stemmed from solely the distortion mentioned above or from the coaction of the distortion and desensitization.) Thus, it can be concluded that the long-term oral CBD treatment significantly augmented the adenosinergic signaling of the heart even if the myocardial A₁ adenosine receptors underwent some downregulation and/or desensitization.

The observation that the presence of CBD in the heart led to A₁ adenosine receptor activation (Gonca and Darici, 2015) has raised the possibility that CBD might act as an A₁ adenosine receptor agonist ⁸ . If so, CBD, persisting in the myocardium of the CBD-treated rats, had to act also as a surplus A₁ adenosine receptor agonist. Based on this, CBD, as a surplus A₁ adenosine receptor agonist, should have decreased the virtual response to both CPA and adenosine (indicated by a conventionally evaluated and plotted E/c curve). It should be noted that a surplus A₁ adenosine receptor agonist (being present even before the E/c curve construction and being neglected) cannot exert any effect that would increase the (real or virtual) response to adenosine (or any other adenosine receptor agonist). However, in our present investigation, the in vivo CBD treatment decreased the response only to CPA, while it increased the response to adenosine (Figure 1). Thus, under our experimental conditions, CBD behaved as an adenosine transport blocker rather than an A₁ adenosine receptor agonist. This result may offer strong evidence in the debate concerning whether the adenosine transport blocker or the adenosine receptor agonist property dominates the effect of CBD on the myocardial adenosinergic signaling.

As a limitation of this study, we should mention that, before being sacrificed, the animals used for this study underwent several in vivo investigations (for more details, see the Materials and Methods), which may have interfered with our measurements to some extent. Furthermore, the CBD-rich hemp extract oil, used for this study, contained other bioactive phytochemicals as well, effects of which may have contributed to our results. According to the manufacturer, this product contains cannabidivarin (0.038% w/w) in addition to CBD (16.802% w/w), and other components may also be present below the limit of quantification ⁹ . Based on results of others, cannabigerol, cannabichromene, Δ⁸-and Δ⁹-tetrahydrocannabinols, cannabinol, cannabicyclol (and their 3′-carboxy derivatives as “acids”), furthermore cannflavins, β-myrcene and β-caryophyllene may be the further components (Morales et al., 2017; Pellati et al., 2018; Ohtsuki et al., 2022). All have antioxidant property, most of them have been reported to exert antiinflammatory effects, and some have also been proposed to be neuroprotective (Morales et al., 2017; Scandiffio et al., 2020; Pannico et al., 2022). Importantly, the CBD-rich extracts appeared to be superior to the purified CBD products at the levels of both their beneficial and undesirable effects (Millar et al., 2020). Finally, the results of the present study rest on the assumption that the in vivo administered CBD was present in a sufficient amount in the atria to inhibit ENT1 during our ex vivo experiments. Our reason to assume this is the highly lipophilic nature of CBD. In our recent study, FSCPX, another highly lipophilic agent, was found to accumulate and persist in the lipid compartment (e.g. cell membranes) of the isolated atria, where it was able to exert its effects for a long time (Viczjan et al., 2021). Thus, it is reasonable to presume that CBD, administered in vivo, can persist in the atria long enough to elicit its effect ex vivo on the transmembranous ENT1.