Introduction

Mitochondrial dysfunction, oxidative damage, and excitotoxicity constitute confluent and interrelated toxic mechanisms triggering neurodegeneration. Combined, disrupted cell metabolism, reactive oxygen species (ROS) formation, and overactivation of the N-methyl-D-aspartate receptor (NMDAr) subtype of glutamate receptors lead to toxic cascades with ensuing excitotoxicity (Muddapu et al. 2020). 3-Nitropropionic acid (3-NP) is a mitochondrial toxin capable of resembling several pathological features of Huntington’s disease in cell and in mammal in vivo models (Túnez et al. 2010; Hariharan et al. 2014; Maya-López et al. 2017). Selective inhibition of succinate dehydrogenase (SDH, electron transport chain complex II) activity, ATP depletion, blockade of the Krebs cycle, increased ROS formation, and secondary excitotoxicity are among the toxic effects produced by 3-NP in the CNS (Brouillet et al. 2005; Chaturvedi and Beal 2013; Maya-López et al. 2017). The propensity of 3-NP to induce the aforementioned effects has been also extended to short-term in vitro studies (Reyes-Soto et al. 2022).

The endocannabinoid system (ECS, or endocannabinoidome) consists of endocannabinoids, receptors, and the enzymes for their synthesis and degradation, and regulates several modulatory processes, including neurotransmission, immune control, and cell signaling. The activation of cannabinoid type 1 (CB1R) and type 2 (CB2R) receptors by endocannabinoids has been suggested to be responsible for the abovementioned effects (Viscomi et al. 2010; Aymerich et al. 2018). The most abundant and physiologically relevant endocannabinoids 2-arachydonoylglycerol (2-AG) and anandamide (AEA) bind to CB1R and/or CB2R, thus acting as major regulators of the central and peripheral nervous systems (Colín-González et al. 2016; Aymerich et al. 2018). The enzymatic hydrolysis of 2-AG by monoacylglycerol lipase (MAGL), and of AEA by fatty acid amide hydrolase (FAAH), is a major physiological event coordinating the end of the endocannabinoid signaling (Cravatt et al. 1996; Dinh et al. 2002; Aymerich et al. 2016). The pharmacological inhibition of MAGL and FAAH affords neuroprotective efficacy by reducing excitability via modulation of cannabinoid signaling (Cravatt et al. 1996; Petrosino and Di Marzo 2010; Carloni et al. 2012; Xu and Chen 2015; Fucich et al. 2020).

Selective inhibition of MAGL activity by JZL184 has been shown to elicit anti-inflammatory, anti-hyperalgesic, and neuroprotective effects (Kerr et al. 2013; Khasabova et al. 2014; Zhang and Thayer 2018). These effects are likely produced by decreasing the formation of beta-amyloid in APdE9 mice and in glial cells (Pihlaja et al. 2015), releasing neuroprotective and anti-inflammatory molecules in a murine MPTP parkinsonian model (Fernández-Suárez et al. 2014). This may also occur in hippocampal neurons via activation of the Nrf2/ARE signaling pathway in a model of oxygen–glucose deprivation (Xu et al. 2021). In this regard, several groups have attributed the protective effects of JZL184 to selective activation of CB2R (Aymerich et al. 2016), which are mainly expressed in immune cells.

Despite neuroprotective effects elicited by the blockade of endocannabinoid hydrolysis have already been described, the protective properties of JZL184 under toxic conditions such as mitochondrial dysfunction, oxidative damage, and excitotoxicity have yet to be demonstrated at different experimental levels. Therefore, in this study, we investigated for the first time the effects of JZL184 on different markers of the short-term 3-NP-induced toxicity in two biological preparations: crude rat brain synaptosomal/mitochondrial fractions—nerve endings containing the whole biochemical machinery for neurotransmission and cannabinoid metabolism—and rat cortical slices—tissue slices comprising the whole cellular diversity of the brain cortex and its integral circuitry—and evaluated the role of CB1R and CB2R using specific antagonists/inverse agonists as a first pharmacological approach for the design of MAGL inhibition-based therapies. Both biological preparations employed in this study represent simple and representative fractions of the brain; they are functional and easy to obtain, thus constituting an optimal first experimental approach in toxicological and pharmacological studies. The 3-NP-induced loss of mitochondrial reductive capacity and oxidative damage were challenged with JZL184, which displayed protective effects.

Material and Methods

Reagents

AM281 (1-(2,4-dichlorophenyl)-5-(4-iodophenyl)-4-methyl-N-4-morpholinyl-1H-pyrazole-3-carboxamide) and JTE907 (N-(1,3-benzodioxol-5-ylmethyl)-1,2-dihydro-7- methoxy-2-oxo-8-(pentyloxy)-3-quinolinecarboxamide) were obtained from Tocris Bioscience (Bristol, UK). JZL184 (4-nitrophenyl 4-[di(2H-1,3-benzodioxol-5-yl)(hydroxy)methyl]piperidine-1-carboxylate), 3-NP (3-nitropropionic acid), thiobarbituric acid (TBA), MTT (3-(4,5-dimethylthizol-2-yl)-2,5-diphenyltetrazolium bromide), D-mannitol, sodium azide, and dichloroindophenol sodium salt were all purchased to Sigma Chemical Co. (Sigma-Aldrich, St. Louis Missouri, USA). Quinolinic acid (QUIN) was obtained from Spectrum Chemical (New Brunswick, NJ, USA). Other reagents were obtained from known commercial sources.

Animals

Adult male Wistar rats (250–300 g; N = 30 (total number of animals used); n = 6 experiments per group; one rat representing one whole experiment) were used for the preparation of synaptosomal/mitochondrial fractions (12 rats, 6 for mitochondrial function assays and 6 for lipid peroxidation assays) and for isolation of cortical slices (18 rats, 6 for mitochondrial function assays, 6 for lipid peroxidation assays, and 6 more for the estimation of the succinate dehydrogenase activity). Since experiments were not carried out in living organisms, but in biological preparations obtained from their dissected brains, each animal served on a given day to obtain the preparations (either synaptosomes or slices). All animals were obtained from the vivarium of the Universidad Autónoma Metropolitana-Iztapalapa, kindly donated by Dr. Socorro Retana-Márquez. Rats were housed at controlled conditions of room temperature (25 ± 3 °C), humidity (50%), and light–dark cycles (12:12 h), with food and water ad libitum. All experimental manipulations were carried out following the guidance of the National Institute of Health Guide for the Care and Use of Laboratory Animals (NIH Publication No. 80–23) revised 1996, and the local Ethical Committees. Formal approval to conduct the experiments described was obtained from the animal subjects review board of the Instituto Nacional de Neurología y Neurocirugía (project numbers 17/20 and 73/20). All efforts were made to minimize animals’ suffering during the experiments.

Treatment of Crude Synaptosomal/Mitochondrial Fractions

Crude synaptosomal/mitochondrial P2 fractions were obtained from adult rat brains, according to a method reported previously by us (Rangel-López et al. 2015; Maya-López et al. 2020). Briefly, the collected brains (without the cerebellum) were homogenized in 10 volumes (g/ml) of sucrose (0.32 M) and centrifuged for 10 min at 1073 g (4 °C); supernatants were recovered and re-centrifuged for 15 min at 17,000 g (4 °C). The obtained pellets were resuspended in 40 volumes (ml) of HEPES buffer (0.1 M NaCl, 0.001 M NaH2PO4, 0.005 M NaHCO3, 0.001 M CaCl2, 0.006 M glucose, 0.01 M HEPES, pH 7.4). Aliquots of equal volumes were briefly stored at −70 °C until used throughout the experiments. Protein content in synaptosomal fractions was estimated by Bradford’s method (Bradford 1976). The collected fractions were incubated in the presence of 3-NP (0.1–1 mM), 100 μM QUIN, or 50 µM FeSO4 for 60 min. Additional fractions were preconditioned for 30 min with increasing concentrations JZL184 (0.1–10 nM), the selective CB1R antagonist/inverse agonist AM281 (10 nM, added 15 min before JZL184), and/or the selective CB2R inverse agonist JTE907 (0.13 nM, also added 15 min before JZL184). Although several reports in the literature have described the toxic concentrations of 3-NP (Reyes-Soto et al. 2022), we proceeded to determine our own concentration-dependent response curve to establish the optimal toxic concentration for our experiments. The toxic concentration of QUIN was chosen from previous reports (Rangel-López et al. 2015; Chavira-Ramos et al. 2021). The concentration range for JZL184 tested herein was inferred from a previous report in which it was demonstrated that 8 nM corresponded to the IC50 for MAGL inhibition in vitro, while 10 µM was optimal to inhibit FAAH (Long et al. 2009). The AM281 and JTE907 concentrations were also inferred from previous reports (Lan et al. 1999; Li et al. 2010; García-Morales et al. 2015; Sánchez-Rodríguez et al. 2018; Chavira-Ramos et al. 2021; Reyes-Soto et al. 2022). All samples were prepared for the experimental assays designed to estimate loss of mitochondrial function and oxidative damage to lipids following their incubation in the presence of the abovementioned treatments. The quantification of protein content in samples was carried out also according to Lowry’s method (Lowry et al. 1951).

Treatment of Cortical Slices

Cortical slices were obtained according to methods previously reported (Colín-González et al. 2014; Reyes-Soto et al. 2022). Cortical slices were collected and sectioned from the frontal cortex of rat brains using a chopper. The slices (~250-µm thickness) were incubated in Krebs–Ringer modified buffer (NaCl 135 mM, KCl 5 mM, MgSO4 1 mM, K2HPO4 0.4 mM, glucose 5.5 mM, HEPES 20 mM, and CaCl2 20 mM) for 30 min at 37 °C in the presence of 5% CO2. Samples were added with JZL184 (10 nM) as a pretreatment for 30 min, and then added with 3-NP (1 mM) and incubated for 60 min at 37 °C and 5% CO2. Some slices treated with JZL184 + 3-NP were also added with AM281 (10 nM) or JTE907 (0.13 nM) for 30 min prior to JZL184. Under the experimental conditions tested in this study, a higher concentration of AM281 (1 µM) decreased the mitochondrial reductive capacity (data not shown), denoting toxicity per se. Experimental groups were categorized as follows: (A) control (an experimental condition incubated only with vehicles for all compounds: H2O instead of AM281 or JTE907; DMSO 1% instead of JZL184; PBS instead of 3-NP); (B) 3-NP alone; (C) JZL184 alone; (D) JZL184 + 3-NP; (E) AM281 + J ZL184 + 3-NP; and (F) JTE907 + JZL184 + 3-NP. The concentrations of all compounds used in this biological preparation correspond to those described above for synaptosomal/mitochondrial fractions.

Functional Assessment of Mitochondrial Activity

The functional status of mitochondrial function was assessed by the MTT reduction assay, according to previous reports (Colín-González et al. 2014; Chavira-Ramos et al. 2021; Reyes-Soto et al. 2022). The treated synaptosomal/mitochondrial fractions and cortical slices were added to 300 µL Krebs’s buffer plus 15 µL MTT reagent (5 mg/ml) and incubated for 60 min at 37 °C. The content of formazan was estimated at 570 nm in a Cytation 3 Imaging Reader (BioTek Instruments, Winooski, VT, USA), and the results were expressed as percent of MTT reduction vs. the control.

Functional Assessment of Oxidative Damage to Lipids

Oxidative damage to lipids was assessed in cortical slices according to previous reports (Colín-González et al. 2014; Reyes-Soto et al. 2022). After treatment, the synaptosomal/mitochondrial fractions and the slices were homogenized in lysis buffer, and 100 µl-aliquots were added to 50 µl of thiobarbituric acid (TBA) reagent (0.75 g TBA + 15 g trichloroacetic acid + 2.53 ml HCl) and incubated in a boiled water-bath for 20 min. Samples were then centrifuged at 3000 g for 10 min at 4 °C. The absorbance was evaluated at 532 nm in a Cytation 3 Imagin Reader (BioTek) in the collected supernatants. Results, originally calculated as nmols of TBA-reactive substances (TBARS) formed per mg of protein, were finally expressed as percent of lipid peroxidation vs. the control.

Functional Assessment of Succinate Dehydrogenase Activity

Succinate dehydrogenase (SDH) activity was quantified in cortical slices according to a method recently reported (Reyes-Soto et al. 2022). Tissue samples were homogenized in buffer solution and centrifuged (3000 g); then, aliquots (400 µL) were incubated in an added solution of phosphate buffer assay reaction (650 µL; pH 7.0) containing 0.3 M d-mannitol + 5.0 mM magnesium soluble chloride + 125 µL of 0.004 M sodium azide + 125 µL of 0.50 mM dichloroindophenol + 125 µL of 0.2 M succinate. The reduction of dichlorophenolindophenol (DCPIP; assessed as the decay of color) was recorded over 40 min at 600 nm. An UV-1603 Shimadzu (Kyoto, Japan) spectrophotometer was used to detect optical density of the solutions. The content of protein in samples was measured according to previous protocols (Lowry et al. 1951), and results were expressed as units of reduction (SDH activity) per milligram of protein (U/mg protein), where one unit of SDH activity is defined as the amount of enzyme capable of reducing 1 µmol DCPIP per min at 25 °C (Popov et al. 2010).

Statistical Analysis

Results depict independent experiments (one experiment per animal) expressed as mean values ± standard error of n = 6 experiments per group. All experiments were performed in duplicate, and mean values were used for statistical calculations. Results were analyzed by one- or two-way analysis of variance (ANOVA) followed by post hoc Dunnett’s or Tukey’s tests (GraphPad, Scientific, San Diego, CA, USA). Values of p ≤ 0.05 were considered of statistical significance.

Results

JZL184 Ameliorated the 3-Nitropropionic Acid-Induced Loss of Mitochondrial Function in Synaptosomal/Mitochondrial Fractions in a CB2R-Related Manner

To establish and confirm the optimal concentrations of 3-NP and JZL184 to be tested, first we evaluated concentration–response effects of these compounds on the basal levels of mitochondrial function in synaptosomal/mitochondrial fractions.

As shown in Fig. 1A, 3-NP produced a concentration-dependent decrease in mitochondrial function in a concentration range between 0.1 and 1 mM, with significant effects at 0.3 (p ≤ 0.05 vs. the control), 0.5, and 1 mM (p ≤ 0.01 vs. the control group). Since 1 mM exerted the strongest effect and this concentration is often used in other reports, we decided to use this concentration from this point on. As shown in Fig. 1B, JZL184 did not affect the basal (control; 0.6 OD units of MTT reduction) mitochondrial function at any of the concentrations tested. From this point on, we used the 10 nM concentration for subsequent experiments.

Fig. 1
figure 1

A, B Concentration-dependent effects of 3-nitropropionic acid (3-NP, in A) and JZL184 (JZL, in B) on basal mitochondrial function (mitochondrial reductive capacity) in rat brain synaptosomal/mitochondrial fractions. C, D Effect of JZL184 (JZL, 10 nM) on the 3-nitropropionic acid (3-NP, 1 mM, in A)- and quinolinic acid (QUIN, 100 µM, in B)-induced loss of mitochondrial reductive capacity in rat brain synaptosomal/mitochondrial fractions. AM281 (10 nM) and JTE907 (0.13 nM) were also used to investigate the role of cannabinoid receptors. Values represent the mean ± one SEM of n = 6 experiments per group. *p ≤ 0.05, **p ≤ 0.01, and ***p ≤ 0.001, different to the control group (CT); &&p ≤ 0.01, different to 3-NP; $$$p ≤ 0.001, different to QUIN; ##p ≤ 0.01, different to JZL+3-NP. For A and B: one-way ANOVA followed by Dunnett’s test. For C and D: two-way ANOVA followed by Tukey’s test

Full size image

Next, we tested the effects of JZL184 on the 3-NP-induced loss of mitochondrial function in synaptosomes. As shown in Fig. 1C, 3-NP decreased mitochondrial function to ~55% vs. the control group (p ≤ 0.001). In turn, mitochondrial activity induced by 3-NP was restored to levels indistinguishable from controls (~90%; p ≤ 0.01 compared to the 3-NP group) upon treatment with JZL184. In synaptosomes preconditioned with JZL184 and exposed to 3-NP, the selective CB2R inverse agonist JTE907 returned mitochondrial activity to levels analogous to 3-NP treatment alone (p ≤ 0.01 compared to the JZL184 + 3-NP group). The CB1R antagonist/inverse agonist AM281 did not significantly alter the effect of JZL184 on 3-NP toxicity. In addition, no significant effect on basal mitochondrial function was noted upon JZL184 treatment.

Figure 1D depicts the effect of JZL184 on QUIN-induced loss of mitochondrial function. The neurotoxic model induced by QUIN was performed for comparative purposes to confirm the protective effects of JZL184. As expected, QUIN was less effective in decreasing mitochondrial activity than 3-NP (~75% vs. the control group; p ≤ 0.001). JZL184 completely prevented the QUIN-induced loss of mitochondrial function, returning mitochondrial activity to basal levels (p ≤ 0.001 compared to the QUIN group); however, in contrast to the 3-NP toxic model, JTE907 did not modify the effect of JZL184 on QUIN toxicity. AM281 had no effect on the effects elicited by JZL184 in this toxic model. Furthermore, JZL184 alone did not modify the basal mitochondrial reductive capacity

JZL184 Prevented the Oxidative Damage to Lipids Induced by 3-Nitropropionic Acid in Synaptosomal/Mitochondrial Fractions in a CB1R-Related Manner

Next, we explored whether JZL184 could prevent 3-NP-induced lipid peroxidation (an index of oxidative cell damage induced by the noxious actions of reactive oxygen species) in synaptosomes as to determine whether redox modulatory effects are inherent to the protection afforded by MAGL inhibition.

As shown in Fig. 2A, 3-NP increased lipid peroxidation by ~20% vs. the control (p ≤ 0.01), while pretreatment with JZL184 decreased the effect of 3-NP to basal levels (p ≤ 0.001 compared to the 3-NP group). AM281, but not JTE907, abolished the protective effect of JZL184 on 3-NP-induced lipid peroxidation (p ≤ 0.001 compared to the JZL184 + 3-NP group). JZL184 did not produce oxidative damage to lipids compared to control values (17 nmol TBARS/mg protein).

Fig. 2
figure 2

Effect of JZL184 (JZL, 10 nM) on 3-nitropropionic acid (3-NP, 1 mM, in A)- and iron sulfate (FeSO4, 50 µM, in B)-induced lipid peroxidation (TBARS formation) in rat brain synaptosomal/mitochondrial fractions. AM281 (10 nM) and JTE907 (0.13 nM) were also used to investigate the role of cannabinoid receptors. Values represent the mean ± one SEM of n = 6 experiments per group. **p ≤ 0.01 and ***p ≤ 0.001, different to the control (CT); &&&p ≤ 0.001, different to 3-NP; %%%p ≤ 0.001, different to JZL + 3-NP; ap ≤ 0.05, different to FeSO4. Two-way ANOVA followed by Tukey’s test

Full size image
Fig. 3
figure 3

Effect of JZL184 (JZL, 10 nM) on 3-nitropropionic acid (3-NP, 1 mM)-induced loss of mitochondrial reductive capacity (in A) and lipid peroxidation (TBARS formation; in B) in rat brain cortical slices. AM281 (10 nM) and JTE907 (0.13 nM) were also used to investigate the role of cannabinoid receptors. Values represent the mean ± one SEM of n = 6 experiments per group. ***p ≤ 0.001, different to the control (CT); &&p ≤ 0.01 and &&&p ≤ 0.001, different to 3-NP. Two-way ANOVA followed by Tukey’s test

Full size image

Figure 2B shows the effect of JZL184 on the FeSO4-induced lipid peroxidation in synaptosomal/mitochondrial fractions. The pro-oxidant effect of FeSO4 was tested as a positive control of lipid peroxidation to evaluate whether the protective profile of JZL184 involved direct antioxidant properties. As expected, FeSO4 increased lipid peroxidation by ~100% compared to the control group (p ≤ 0.001), whereas JZL184 not only was unable to reduce the effect of FeSO4, but also increased oxidative damage in synaptosomes (p ≤ 0.05 compared to the FeSO4 group).

JZL184 Ameliorated the 3-NP-Induced Loss of Mitochondrial Function and Lipid Peroxidation in a CB1R- and CB2R-Independent Manner

To confirm the protective effect of JZL184 in a more complex biological preparation containing multiple types of cells with an integral circuitry and well-preserved cytoarchitecture, the effects of JZL184 on 3-NP-induced mitochondrial dysfunction/loss of cell viability and lipid peroxidation were tested in fresh cortical slices.

As shown in Fig. 3A, 3-NP decreased the basal mitochondrial reductive capacity by ~50% compared to the control group (p ≤ 0.001), while JZL184 attenuated this effect by ~80% (p ≤ 0.01 compared to the 3-NP group alone). Neither AM281, nor JTE907, modified the effect of JZL184 on the 3-NP-induced loss of mitochondrial function/cell viability. JZL184 did not modify the basal levels of mitochondrial function (control; 0.73 OD units of MTT reduction).

As shown in Fig. 3B, 3-NP increased lipid peroxidation by ~95% compared to the control group (p ≤ 0.001), while JZL184 attenuated this effect to basal levels (p ≤ 0.001 compared to the 3-NP group). Again, neither AM281, nor JTE907, modified the effect of JZL184 on the 3-NP-induced oxidative damage to lipids. JZL184 did not modify the basal levels of lipid peroxidation (control; 1.15 nmol TBARS/mg protein).

JZL184 Recovered the 3-NP-Induced Inhibition of SDH Activity

We also explored the effect of JZL184 on 3-NP-induced inhibition of SDH activity (Fig. 4). 3-NP alone decreased the levels of SDH activity by 65% compared to the control (p ≤ 0.0001), whereas JZL184 attenuated this effect by 95% compared to the 3-NP group (p ≤ 0.001, different to 3-NP). JZL184 alone did not change the baseline SDH activity (control; 17.5 units of SDH activity).

Fig. 4
figure 4

Effect of JZL184 (JZL; 10 nM) on 3-nitropropionic acid (3-NP, 1 mM)-induced succinate dehydrogenase (SDH) decreased activity in cortical slices. Values represent the mean ± one SEM of n = 6 experiments per group. ****p ≤ 0.0001, different to the control (CT); &&&p ≤ 0.001, different to 3-NP. Two-way ANOVA followed by Tukey’s test

Full size image

Discussion

2-AG is present in several brain regions, including the cerebral cortex, hippocampus, hypothalamus, diencephalon, striatum, mesencephalon, and cerebellum (Sugiura et al. 2006). It is also found in the spinal cord and the hypophysis. This endocannabinoid is synthesized on demand by the action of diacylglycerol lipase at the cytoplasmatic and nuclear membrane levels (García del Caño et al. 2015) and has been shown to evoke neuroprotective effects through the activation of both CB1R and CB2R, being CB1R the main regulator of neurotransmitters release as it is abundantly found in presynaptic terminals of the substantia nigra, striatum, hippocampus, cerebellum, and cerebral cortex (Kawamura et al. 2006; Tsuboi et al. 2018). The main enzyme responsible for 2-AG hydrolysis is MAGL (Scalvini et al. 2016), which has recently emerged as a possible therapeutic tool, as it can control not only the levels of 2-AG, but also of other monoacylglycerides, and indirectly, free fatty acids, including lipids with proinflammatory effects (Gil-Ordóñez et al. 2018). MAGL is localized in cytoplasmatic membranes as well as in the membrane of the endoplasmic reticulum and is equally distributed among the cytosolic and membrane fractions (Grabner et al. 2017). Notably, the genetic and pharmacological inhibition of MAGL has been shown to increase 2-AG levels, leading to desensitization of CB1R (Chanda et al. 2010; Schlosburg et al. 2010; Navia-Paldanius et al. 2015). In a chronic pharmacological model, mice treated with JZL184 (40 mg/kg/day i.p.) for six consecutive days showed a significant decrease in the number and function of CB1R (Schlosburg et al. 2010). In addition, in a MAGL-KO model, mice exhibited a compensatory response in CB1R signaling in brain areas associated with anxiety (Imperatore et al. 2015). Altogether this evidence suggests that 2-AG is responsible for the negative regulation of CB1R, it should be considered that these studies were carried out under in vivo and chronic conditions. In contrast, our findings suggest that under in vitro and short-term conditions, at least in synaptosomal fractions, the inhibition of MAGL recruits the activation of CB2R for protection against the loss of mitochondrial function, and CB1R for protection against oxidative damage, likely due to a direct action of 2-AG on these receptors, and this conclusion was drawn upon the use of the selective CB1R antagonist/inverse agonist AM281 and/or the selective CB2R inverse agonist JTE907, two pharmacological tools capable of revealing the involvement of specific cannabinoid receptors in paradigms involving cannabinoid signaling (Pertwee 2010), like the one explored in this report.

Short-term studies are relevant to identify primary targets for compounds in biological preparations containing receptors and signaling molecules that regulate cytoprotective responses. In this novel report, we demonstrate for the first time that selective inhibition of MAGL by JZL184 affords protection against mitochondrial dysfunction and oxidative damage to lipids induced by 3-NP in two biological preparations, both containing the biochemical machinery to metabolically process endocannabinoids and their receptors. These protective effects were confirmed also in the excitotoxic model produced by the NMDA agonist, QUIN. This pharmacological strategy was effective in preventing the loss of mitochondrial reductive capacity induced by 3-NP and QUIN, as well as the 3-NP- and FeSO4-induced lipid peroxidation in synaptosomal/mitochondrial fractions, both in a cannabinoid receptor-dependent manner. Notably, JZL184 also restored mitochondrial reductive capacity, decreased the oxidative damage to lipids, and prevented the loss of SDH activity induced by 3-NP in cortical slices in a cannabinoid receptor-independent manner. This broad spectrum of protective effects, as well as the differential involvement of cannabinoid receptors, suggests wide efficacy for JZL184 as a therapeutic tool, while recognizing that its mechanisms of action need more detailed characterization. In previous reports, we assessed the efficacy of URB597, a selective inhibitor of FAAH, in protecting cellular fractions and murine models against the toxic effects induced by 3-NP and QUIN by CB1R- or CB2R-dependent mechanisms (Maya-López et al. 2017; Aguilera-Portillo et al. 2019; Chavira-Ramos et al. 2021); however, MAGL inhibition and further 2-AG accumulation offers a more promising strategy as 2-AG is the most abundant endocannabinoid in the CNS, and is a full agonist for cannabinoid receptors, also holding several other targets for a broader regulation of physiological responses (Sugiura et al. 2006; Aymerich et al. 2016; Zhu et al. 2023). One weakness of our study is the lack of analysis of 2-AG levels in our preparations, as it may provide unequivocal evidence for the endocannabinoid’s role in the protective effects described in this report. Nonetheless, there is a considerable amount of evidence on the induction of 2-AG levels by MAGL inhibition, but this approach should be considered in further studies.

In turn, direct evidence on the neuroprotective effects of JZL184 has been reported under different experimental conditions. JZL184 was shown to reduce amyloid beta-induced neurodegeneration and apoptosis in hippocampal neurons in culture by increasing the levels of 2-AG, which acted via CB1R-dependent suppression of extracellular signal-regulated kinases 1 and 2 (ERK1/2) and nuclear factor-κB (NF-κB) phosphorylation and cyclooxygenase-2 (COX-2) expression (Chen et al. 2011). The exogenous application of 2-AG and/or the increase in endogenous 2-AG levels by JZL184 abolished the interleukin-1β- and lipopolysaccharide-induced reduction of PPARγ expression in mouse hippocampal neurons in culture, suggesting that these receptors may be involved in the protective properties of 2-AG (Du et al. 2011). In addition, cerebral MAGL activity, via an alternative pathway, has been shown to promote neuroinflammatory prostaglandin production through the hydrolysis of 2-AG, thus highlighting the importance of MAGL inhibition as an experimental strategy to reduce neuroinflammation and neurodegeneration (Nomura et al. 2011). More recently, JZL184 was shown to improve cell survival in SH-SY5Y neuroblastoma dopaminergic neurons in a CB2R-dependent mechanism (Aymerich et al. 2016). Notably, these protective effects were produced at a concentration much higher (100 nM) that the one we used herein (10 nM), which is closer to the IC50 reported for JZL184 in mice (8 nM); this condition was justified by the fact that SH-SY5Y are human neuroblastoma cells, therefore being less sensitive than murine cells. Moreover, although these authors suggested that neuroprotection induced by JZL184 might be selective to dopaminergic cells, as it failed to protect cells with striatal phenotype, here we have demonstrated for the first time that this compound affords neuroprotection in toxic models typically linked to Huntington’s disease by mechanisms dependent and independent of cannabinoid receptors. In further support to our findings that protective effects of JZL184 do not necessarily involve CB2R activation, recent evidence by Zhang and Chen (2018) suggested that the expression of APP and β-secretase, as well as production of total Aβ and Aβ42, was significantly reduced in APP transgenic mice lacking CB2R (TG-CB2-KO) treated with JZL184, thus suggesting the activation of alternative protective mechanisms by this compound. Furthermore, the protective effects in vivo of JZL184 have been extended to the attenuation of infarct volume and hemispheric swelling in a model of experimental ischemic stroke in rats via inhibition of inflammatory responses and eicosanoid production rather than activation of cannabinoid receptors (Choi et al. 2018), thus supporting the concept of alternative mechanisms for JZL184-induced neuroprotection. In line with this evidence, in hippocampal neurons in culture exposed to oxygen/glucose deprivation/reoxygenation representing an in vitro model of cerebral ischemia/reperfusion, JZL184 was shown to improve cell viability, inhibit ROS formation, increase antioxidant defense, and decrease proinflammatory cytokines (Xu et al. 2021). These authors suggested that direct activation of the antioxidant Nrf2/ARE signaling pathway was responsible for the neuroprotective responses observed. Indeed, this specific mechanism may help to explain the differential protective mechanisms observed in our study, as activation of the Nrf2/ARE pathway may occur in cortical slices conformed by integral cell bodies (Reyes-Soto et al. 2022), whereas activation of cannabinoid receptors is more likely to occur in synaptosomal fractions, a biological preparation lacking most of the components of the Nrf2/ARE pathway. Thus, additional studies are warranted to investigate whether JZL184 selectively activates one or more of the aforementioned mechanisms to induce these protective responses.

Emphasis should also be directed to the fact that CB1R and CB2R are not the only receptors activated by 2-AG, as it has been demonstrated that this endocannabinoid can also act on nuclear peroxisome proliferator-activated receptor-gamma (PPARγ) (Pertwee et al. 2010; Du et al. 2011), which is an important mediator in signaling against neuroinflammation (Xu and Chen 2015), thus representing a potential alternative mechanism to explain the effects observed in cortical slices. In addition, 2-AG binds and activates the member 1 of subfamily V of transient receptor potential vanilloid 1 (TRPV1) cationic channel both in vitro and in vivo (Di Marzo 2018; Cristino et al. 2020), as well as GABA receptors and G protein-coupled receptor 55 (GPR55), thereby regulating the activity of excitatory hippocampal neurons (Sylantyev et al. 2013). The lack of effect of the CB1R and CB2R antagonists/inverse agonists found in cortical slices suggests that in contrast to what was observed in synaptosomal/mitochondrial fractions, MAGL inhibition in slices might induce preferential activation of non-classical cannabinoid receptors, such as PPARγ, the most plausible candidates; therefore, our study should be extended to explore the role of both PPARγ and TRPV1 receptors using the same toxic paradigm. Moreover, short-term 2-AG accumulation noted in our study should retain the endocannabinoid preferentially at the intracellular level, where it can reach these organelle receptors. In addition, at this point, we cannot discard the possibility that some of the effects reported herein might also derive from the transformation of 2-AG in prostaglandin-glyceryl esters by COX-2 (Hu et al. 2008), which adds a possible key mechanism to cannabinoid receptor-independent effects. Of further consideration, decreased 2-AG degradation by MAGL inhibition also means lower generation of arachidonic acid from 2-AG, which may also affect the transformation of this fatty acid into different eicosanoids (Yui et al. 2015), thus displaying several other pharmacological effects needing further investigation. These alternative mechanisms are particularly relevant for the toxic model induced by QUIN, where neither CB1, nor CB2 blockade modified the protective effect of JZL184. Probably, in this excitotoxic model, JZL184 exerts its protective effects preferentially through cannabinoid receptor-independent mechanisms, this finding deserving further investigation.

Based on these considerations, JZL184-induced pro-energetic, anti-inflammatory, and antioxidant effects elicited by MAGL inhibition, and the subsequent 2-AG accumulation might contribute to the protective responses against the toxic effects evoked by 3-NP and QUIN. Finally, it is clear that, either dependently or independently of cannabinoid receptors, the preservation of mitochondrial reductive capacity and the reduction of oxidative damage are both early protective processes that should be linked to the preservation of SDH activity, which in turn display sufficient efficacy to protect the biological preparations assessed in this study.

Concluding Remarks

Our novel findings support and confirm the concept that the modulation of the ECS focused on the accumulation of endocannabinoids constitutes a valuable strategy to prevent the toxic effects of neurotoxic compounds triggering degenerative events. Specifically, JZL184 represents an experimental tool useful to ameliorate the early toxic effects of the mitochondrial toxin 3-NP and the excitotoxin QUIN, in a mechanism involving the inhibition of MAGL and the subsequent accumulation of 2-AG. While both CB1R and CB2R participate in a differential manner in the protective mechanisms exerted by JZL184 in synaptosomal/mitochondrial fractions, none of these receptors seem to be responsible for its protective effects in cortical slices, suggesting that other mechanisms might be involved in protecting these biological preparations. The collected evidence herein presses for more detailed investigations on the protective mechanisms evoked by JZL184 and suggests the design of therapies against neurodegenerative disorders based on agents aimed to reduce endocannabinoids’ hydrolysis.