In this work, we evaluated the protective effects elicited by cannabinoid agents and the possible involvement of mitCB1R in a toxic model induced by 3-NP in isolated rat brain mitochondria. For this purpose, we used the endocannabinoid AEA and the synthetic cannabinoid WIN 55212–2, both well-known CB1R agonists. We also used AM281 as a pharmacological tool to inhibit the activity of mitCB1R. The mitochondrial reduction capacity, ROS formation, and mitochondrial swelling were all quantified to assess the protective capacity of these compounds.

To assess mitochondrial reduction capacity at the level of Complex II, succinate was used as substrate/electron donor for SDH, whereas rotenone was used as a well-known Complex I inhibitor to induce succinate oxidation. Reduction capacity was significantly decreased by 3-NP, while AEA and WIN 55212–2 partially prevented this effect, and this protection was likely to be mediated by mitCB1R, as evidenced by the effect of AM281. Noteworthy, in additional experiments in which mitochondria were incubated in the absence of rotenone, the cannabinoid-induced protective effect was complete, suggesting that, despite the toxic model employed specifically affects Complex II, an active role of Complex I is likely essential for the preservation and integral recovery of mitochondrial activity in 3-NP-exposed mitochondria.

In turn, AEA prevented the 3-NP-induced H₂O₂ formation in mitochondria through a protective mechanism likely to be dependent on mitCB1R activation, suggesting that, in this experimental model, the increased ROS formation induced by 3-NP might be inhibited by mitCB1R activation probably through the regulation of H₂O₂ formation from the mitochondrial respiratory chain; therefore, it is likely that mitCB1R participates in the modulation of ROS formation. In addition, our results show that mitochondrial swelling induced by 3-NP was prevented by AEA probably through the activation of mitCB1R, suggesting that this receptor may modulate the activity of mPTP via the regulation of Ca²⁺ channels.

Regarding the toxic model used in this study, 3-NP is a mycotoxin derived from the fungus species Aspergillus sp. and Artrinium sp., and plants such as Astragalus sp. and Indigofera endecaphylla. This toxin inhibits selectively the mitochondrial Complex II and has been employed to reproduce the pathological features of HD in animal models in vivo, as it triggers mechanisms of neural damage such as mitochondrial dysfunction, excitotoxicity, oxidative stress, inflammation, and neural cell death in the striatum, brain cortex, globus pallidus, and hippocampus, thus generating motor alterations in rodents [44, 45]. Altered striatal cell morphology has been observed in adult male rats lesioned with 3-NP, generating pyknosis, swelling, and striosomal alterations [46]. In this study, we used the 3-NP model in brain-isolated mitochondria to investigate, in a selective manner, the role of Complex II on mitochondrial dysfunction and cannabinoid-mediated protection as it has been demonstrated that this model can trigger toxic signals in vitro that involve mitochondrial dysfunction [17].

Hereby, the use of substrates and inhibitors is relevant to stimulate functional responses in mitochondria isolated from cells obtained from the whole brain, as it is known that mitochondrial function is dependent on both the cell environment and oxidation of biomolecules to provide electron carriers to maintain energy metabolism and ETC. In this context, rotenone is useful in preventing the reverse electron transfer and blocking the accumulation of oxaloacetate, thus inhibiting mitochondrial Complex I [43], thereby augmenting ROS formation in neural cells, in contrast to limited production in astrocytes [47]. Consequently, rotenone is a valuable tool to isolate the function of Complex II by preventing the return of electrons to Complex I.

In turn, MTT, and the subsequent formation of formazan, is a technique that has been used as an index of metabolic activity, proliferation, and cell viability in cell cultures and tissue homogenates [48, 49]. In isolated mitochondria, this method has been associated with the mitochondrial reduction capacity in which complexes from the respiratory chain provide their electrons for MTT reduction, and this effect is dependent of dehydrogenases through cofactors such as NADH, FADH, and NAD(P)H, which catalyze the reaction. Therefore, mitochondrial activity has been verified with inhibitors such as rotenone [49, 50]. Our results show that the reduction capacity generated by SDH in the presence of succinate and rotenone was fourfold higher than that of mitochondria in basal conditions, highlighting the relevance of substrates and inhibitors. Notably, when mitochondria were incubated in the absence of rotenone, the mitochondrial reduction capacity was ninefold higher than that of control mitochondria in basal conditions, demonstrating that Complex I plays a crucial role in regulating the reduction capacity in the toxic paradigm evoked by 3-NP, even though this toxin selectively inhibits Complex II.

The partial protective effect evoked by cannabinoids AEA and WIN 55212–2 on the 3-NP-induced Complex II dysfunction was likely to be mediated by activation of mitCB1R, suggesting that mitCB1R completely coordinates the protective function on mitochondrial activity. In this regard, it is known that only 10% of the mitochondrial reduction capacity corresponds to the Complex II activity [50], which in light of our results, reveals the moderate level of contribution of Complex II to this paradigm. Our results also suggest that mitCB1R activation by cannabinoids under conditions of Complex I inhibition only exerts a partial protective effect on the 3-NP decreased reduction capacity; however, when this effect is explored under conditions in which Complex I remains active (without rotenone), the mitCB1R activation-induced protection is complete, suggesting that the activity of Complex I is crucial to orchestrate protective responses, highlighting the possible role of mitCB1R on the regulation of mitochondrial dehydrogenases. Thus, mitCB1R activation under conditions of an active Complex I is likely to generate complete protection against the toxic effect of 3-NP, supporting the concept that this receptor may regulate the functions of Complex I [31, 32]; however, our results suggest that the activity of Complex II might also be regulated by this receptor, either by direct or indirect mechanisms, through the modulation of mitochondrial ion channels [51]. Moreover, it should be considered that results described by other groups demonstrating the regulation of Complex I by mitCB1R have been collected exploring the physiology of this receptor through the assessment of oxygen consumption in mouse mitochondria from hippocampal neurons under basal conditions [31, 32], whereas our study explores the role of this receptor also under pathological conditions in mitochondria isolated from the whole rat brain. This is particularly relevant as it has been demonstrated that mitochondria from neuronal cells and astrocytes exhibit differential functional features [47], where the first are more sensitive to oxidative stress.

It has also been demonstrated that mitCB1R, which is located in the outer membrane of mitochondria, plays a central role in mouse hippocampal neurons treated with the cannabinoid agents WIN 55212–2 and delta9-tetrahydrocannabinol (Δ9-THC) at nanomolar concentrations, producing a reduction in mitochondrial respiration, Complex I activity, protein kinase A (PKA), and cAMP activities, all under physiological conditions. Combined, these effects contribute to the suppression of endocannabinoid-dependent inhibition of depolarization [32]. This regulatory mechanism is crucial for hippocampal physiology through the modulation of neural energy metabolism and the control of cognitive functions; however, in contrast to these effects, our experiments provide a more complete assessment as they explore the effects of cannabinoids and an approach to the possible role of mitCB1R in mitochondria from the whole brain under toxic conditions, using a plethora of endpoints such as assessment of the reduction capacity, ROS formation, and mitochondrial swelling. Accordingly, our study investigates the integrated response of these receptors under pathological conditions, revealing their novel function. Furthermore, a link between mitochondrial activity and the process of memory formation [31] further demonstrated that bioenergetics processes linked to mitCB1R regulate cognitive functions such as memory, which was concluded by demonstrating that these receptors modulate the activity of Complex I through Gαi/o signaling, as well as PKA and adenylate cyclase (AC) inhibition. Although these important findings on the role of the receptor describe major aspects of neural physiology, they impart limited information on the function of neural mitochondria, while our findings provide information on the protective effects evoked by mitochondria obtained from different brain cell types and their responses to a toxic insult.

Although ROS, such as H₂O₂, serve as second messengers with fundamental signaling activity in physiological processes such as cell proliferation, mitophagy and apoptosis, synaptic plasticity, and ion channel regulation [5], a deficient regulation in redox homeostasis due to toxic conditions is responsible for the onset and development of neurodegenerative events [52]. Therefore, the relevance of maintaining an optimal mitochondrial energy metabolism lies in the fact that most of ROS formation originated in this organelle. In this work we also evaluated H₂O₂ formation induced by 3-NP through the oxidation of DCF. H₂O₂ is generated by the reduction of O₂^·− through the activity of manganese superoxide dismutase [42]. In this context, we determined H₂O₂ formation produced by selective dysfunction of Complex II in the presence of succinate and antimycin as inhibitors of III_Qi site at Complex III to induce reverse transport. Reverse transport directs electrons to Complex I at IQ (the site for ubiquinone reduction), which is the primary origin for O₂^·−, reaching IF (site of flavin) where NADH is oxidized to form O₂^·− [53]. The significant increase in H₂O₂ induced by 3-NP was decreased by AEA in a likely dependent mitCB1R activation manner, suggesting that regulation of Complex I by this receptor might be responsible for the modulation of ROS formation. We advance this hypothesis based on observations that the toxic formation of ROS was mitigated in this model in a receptor activation-mediated manner, and not by the properties of AEA as a free radical scavenger. Therefore, we hypothesize that this mechanism occurs when, following mitCB1R activation by AEA, inhibition of AC may decrease Complex I activity, inhibiting its subunits [54], among which the site 2 of Fe-S of NADH ubiquinone dehydrogenase (NDFUS2) containing the prosthetic group FeS bound to the reduction site of quinone IQ, which when inhibited, will be unable to provide the electrons needed for O₂^·− production, thus preventing its subsequent reduction to H₂O₂, ultimately resulting in a general decrease in ROS formation [55].

Synthetic cannabinoids such as CP55940 and JWH-015 have been shown to exert protective effects against paraquat (Complex I inhibitor) by decreasing ROS formation in rat brain cortical mitochondria due to their properties as free radical scavengers through the presence of hydroxyl groups. These effects were accompanied by attenuation in the membrane potential decrease, also preventing mitochondrial swelling and mitochondrial damage [37]. Despite several cannabinoid compounds, including AEA, have been shown to possess scavenging properties, the fact that mitCB1R activation completely inhibited the ROS formation in this study suggests that mechanisms such as Complex I modulation are more responsible for this protective effect.

Transitory mitochondrial swelling occurs by changes in the volume of mitochondrial matrix and it can be considered a physiological condition for the adequate functioning of mitochondria, as it maintains the ionic concentrations of the cytoplasmic environment, the ETC, and oxidative phosphorylation. The mitochondrial volume is dependent on the fluxes of K⁺ and Ca²⁺ ions through their respective channels, which are located in the inner mitochondrial membrane [56]. The irreversible increase in mitochondrial volume leads to pathological swelling associated with the mPTP opening by loss of inner membrane integrity, triggering changes in mitochondrial morphology in a mechanism linked to the onset of neurodegenerative disorders [17, 56]. However, its role remains poorly understood under physiological conditions, although it has been suggested that it can contribute to the short-term regulation of homeostasis. mPTP opening occurs in response to mitochondrial depolarization induced by Ca²⁺ overload and high levels of ROS formation, and this process uncouples ATP synthase while hydrolyzing ATP, allowing the entrance of ions, water, and solutes with a molecular weight below 1500 Da [57]. 3-NP (5 mM) generates a model of mitochondrial dysfunction by swelling and autophagy, which leads to cell death of SH-SY5Y cells [17]. It has also been tested in rat cortical slices (3-NP 100 µM), where it decreased the mitochondrial reduction capacity while generating cell damage [58]. The decrease in the optical density during the assessment of mitochondrial swelling depends on light transmission, which is proportional to the increase in osmotic pressure following the expansion of the mitochondrial matrix [43, 56]. Given that AEA completely prevented the mitochondrial swelling induced by 3-NP in a likely mitCB1R-dependent manner, we suggest that activation of this receptor might be responsible for modulating mPTP opening through the Gai/o-dependent AC pathway, since this mechanism has been proposed to modulate Ca²⁺ and K⁺ channels [59].

In addition, it has been shown that the synthetic cannabinoids CP55940 and JWH-015 protect cortical rat mitochondria exposed to paraquat [37], where AEA reduced the Ca²⁺-induced cytochrome C release [36]. Our findings support the described evidence, and suggest the involvement of a cannabinoid signaling in the prevention of mitochondrial swelling, though our results suggest that this signal may be mediated by activation of mitCB1R and this process can be linked to the prevention of ROS formation and the subsequent preservation of the mitochondrial reduction capacity.

Our study also suggests that CB1R signaling is analogous in the cytoplasmic and mitochondrial membranes (inhibition of AC and possible regulation of Ca²⁺ and K⁺ channels) [59]. The activation of mitCB1R might regulate the activity of Ca2 + transport, favoring the decrease of Ca²⁺ influx and preventing the Ca²⁺-dependent mPTP opening, thus providing a novel potential mechanism of mitochondrial protection.

Of major relevance is the fact that in vivo evidence supports the concept that CB1R activation by cannabinoids affords protection to the CNS against the alterations induced by 3-NP. The intrastriatal lesion with this toxin in rats decreased the expression of mRNA for CB1R [60]. In CB1R KO mice (the N171-82Q strain), the striatal lesion with 3-NP worsened motor performance and pathological damage of brain tissue [61]. In addition, the use of the selective CB1R agonists arvanil improved motor activity in 3-NP–lesioned rats [62]. In turn, WIN 55212–2 (i.p.) prevented motor alterations, oxidative stress, neural damage, and loss of CB1R expression in rats intrastriatally lesioned with 3-NP [46]. Moreover, the i.p. administration of a well-known inhibitor of fatty acid amide hydrolase (FAAH, the degrading enzyme for AEA), URB597, to rats prevented motor alterations and cell damage in the same toxic model [63]. As a whole, this evidence provides solid support to our present in vitro findings at the mitochondrial level, and generates an important physiological and physiological correlate between the results of this study and those collected from in vivo studies.

Finally, in this study, we used 500 nM of AEA and WIN 55212–2 for most of our experiments, and these concentrations were obtained from our own response-effect curves. While it is known that AEA Ki for CB1R is 61 nM and WIN 55212–2 Ki for CB1R is 62.3 nM [64], we rather used the optimal concentrations of these compounds found in our experimental conditions as little is known yet about the specific properties of mitCB1R. It should be considered that no other forms of cannabinoid receptors have been reported in mitochondria, while these agents have been commonly used to investigate CB1R responses [31, 32]. Our data are also supported by the effect exerted by the selective CB1R antagonist AM281, tested herein at a concentration near its Ki (12 nM) [65], with optimal concentration also obtained from our own concentration curve (Fig. 3A).

Combined, the findings of this study provide new insight into the role of the ECS in the regulation of brain mitochondrial activity under physiological and pathological conditions. Hence, future studies should address whether targeted manipulation of this pathway may afford therapeutic strategy for neurodegenerative diseases.