Introduction

Mitogen-activated protein (MAP) kinases are ubiquitous cell regulatory enzymes, which are grouped into three families: extracellular signal-regulated kinases 1/2 (ERK1/2), c-Jun NH₂-terminal protein kinases 1/2 (JNK1/2) and p38 kinases [1]. In particular, ERK1/2 cascades [2–4] are initiated by the phosphorylation of their upstream kinases MAP kinases kinase/ERK1/2 [MEK1/2 (i.e., ERK kinases [3, 4])] leading to the sequential basal activation of phosphorylation-ERK1/2 (p-ERK1/2), the only known substrates of p-MEK1/2 enzymes [1, 3–5].

The endocannabinoid system consists of two major types of G-protein (Gi/o)-coupled receptors, the cannabinoid 2 (CB₂) and CB₁ receptors that are stimulated by the endogenous ligands anandamide and/or 2-arachidonoylglycerol [6]. In the context of MEK-ERK signaling, the CB₁ receptor agonist WIN55212-2 was shown to induce, as expected, the parallel upregulation of p-MEK1/2 and p-ERK1/2 in rat and mouse cerebral cortex [7, 8]. Moreover, the non-selective CB_1/2 receptor agonist CP55940 was also reported to stimulate p-ERK1/2 in transfected Chinese hamster ovary (CHO) cells expressing the human CB₂ receptor (CHO-CB₂), through pertussis toxin (Gi/o protein-dependent) mechanisms [9].

Notably, the CB₂ receptor (whose expression was initially believed to be limited to peripheral immune cells with low levels in the brain, mainly in microglia) was also recently shown to display constitutive activity (i.e., tonic control or ligand-independent receptor activity [10]), both in CHO-CB₂ membranes [11, 12] and in mouse brain tissue (glia and neurons under physiological conditions, see [13]) which allowed the assessment of various inverse agonists (AM630, JTE 907, and raloxifene) on the in vivo signaling of pro-apoptotic JNK1/2 MAP-kinases [13]. Thus, an inverse agonist at CB₂ receptors is a ligand (e.g., AM630) that exerts the opposite pharmacological effects to a full receptor agonist (e.g., JWH133 [13]).

However, little is known about the sequential in vivo regulation of brain p-MEK1/2 to p-ERK1/2 by mechanistically different CB₂ receptor ligands, i.e., direct agonists, inverse agonists, and endogenous agonists. Therefore, the aim of this study was to quantify the acute effects of CB₂ receptor selective agonists (JWH133), the sustained effects of endogenous agonists (acting on overexpressed receptors) and the acute effects of inverse agonist/antagonist (AM630) on p-MEK to p-ERK signaling in mouse brain cortex.

Materials and methods

Results

CB₂ receptor agonist JWH133 disrupts the sequential regulation of p-MEK1/2 to p-ERK1/2 in the mouse brain cortex

The acute treatment of mice with the CB₂ agonist JWH133 (combined data for 1 mg/kg and 3 mg/kg, i.p., 1 h) modestly reduced, compared with saline controls, the basal activation of p-MEK1/2 in brain cortex (p-MEK/total-MEK: –17% ± 3.7%, n = 8; t = 1.69, P = 0.12) (Figure 1A); data for 1 mg/kg: –14% ± 5.1%, n = 4; t = 1.04, P = 0.34; data for 3 mg/kg: –19% ± 5.6%, n = 4; t = 1.42, P = 0.21. One-way ANOVA for JWH133 (MEK, three groups: F [2, 9] = 1.41, P = 0.29).

Display full size

Figure 1. 

CB₂ receptor ligands disrupt the sequential phosphorylation from MEK to ERK in the mouse brain cortex. A. MEK-ERK activation after acute treatment with the CB₂ agonist JWH133; B. MEK-ERK activation on mice overexpressing the receptor CB₂ (CB₂xP); C. MEK-ERK activation after acute treatment with the CB₂ inverse agonist/antagonist AM630. (A, left panel) Regulation by the CB₂ receptor full agonist JWH133 (1 mg/kg and 3 mg/kg, i.p.) of p-MEK1/2 (p-Ser218/222 MEK/total-MEK ratio) and p-ERK1/2 (p-Tyr204/Thr202 ERK/total-ERK ratio) activities in the mouse brain cortex. Grey (MEK) and white (ERK) columns are means ± SEM (vertical bars) of 8 experiments (with an animal per experiment), and expressed as a percentage of saline (4–5 controls)-treated mice (100%). ^*** P = 0.0009 vs. the corresponding control (unpaired, two-tailed t-test). (A, right panel) Representative immunoblots of cortical MEK (p-MEK and total-MEK) and ERK (p-ERK and total-ERK) form after treatments of mice with the agonist JWH133 (1 mg/kg and 3 mg/kg) or with saline control (C). The apparent molecular masses of MEK and ERK kinase forms were determined by calibrating the blots with prestained molecular weight markers (kDa). (B, left panel) Effects of CB₂ receptor overexpression (38%) on the regulation of p-MEK1/2 (p-Ser218/222 MEK/total-MEK ratio) and p-ERK1/2 (p-Tyr204/Thr202 ERK/total-ERK ratio) activities in mouse brain cortex. Grey (MEK) and white (ERK) columns are means ± SEM (vertical bars) of 9–10 experiments (with an animal per experiment), and expressed as a percentage of WT (10 controls) mice (100%). ^* P = 0.02, ^*** P = 0.0001 vs. the corresponding WT control (unpaired, two-tailed t-test). (B, right panel) Representative immunoblots of cortical MEK (p-MEK and total-MEK) and ERK (p-ERK and total-ERK) form after CB₂ receptor overexpression in mice (WT, CB₂ overexpression). The apparent molecular masses of MEK and ERK kinase forms were determined by calibrating the blots with prestained molecular weight markers (kDa). (C, left panel) Regulation by the CB₂ receptor inverse agonist AM630 (0.3 mg/kg and 1 mg/kg, i.p.) of p-MEK1/2 (p-Ser218/222 MEK/total-MEK ratio) and p-ERK1/2 (p-Tyr204/Thr202 ERK/total-ERK ratio) activities in mouse brain cortex. Grey (MEK) and white (ERK) columns are means ± SEM (vertical bars) of 8–9 experiments (with an animal per experiment), and expressed as a percentage of saline-treated (controls 7–10) mice (100%). ^* P = 0.02 vs. the corresponding control (unpaired, two-tailed t-test). (C, right panel) Representative immunoblots of cortical MEK (p-MEK and total-MEK) and ERK (p-ERK and total-ERK) form after treatments of mice with the inverse agonist AM630 (0.3 mg/kg and 1 mg/kg, i.p) or with saline control (C). The apparent molecular masses of MEK and ERK kinase forms were determined by calibrating the blots with prestained molecular weight markers (kDa)

In contrast to MEK, the basal activation of p-ERK1/2 was significantly upregulated in the same cortical samples (p-ERK/total-ERK: 25% ± 3.6%, n = 8; t = 4.53, P = 0.0009) (Figure 1A); data for 1 mg/kg: 20% ± 5.7%, n = 4; t = 3.04, P = 0.02; data for 3 mg/kg: 30% ± 3.7%, n = 4; t = 5.45, P = 0.001. One-way ANOVA for JWH133 (ERK, three groups: F [2, 10] = 12.2, P = 0.002).

Therefore, the ratio of activity from MEK to ERK quantified in saline controls (ratio of kinases: 0.94) was remarkably reduced (34%) by the CB₂ agonist JWH 1 mg/kg and 3 mg/kg (ratio of kinases: 0.62) (Figure 1A).

Discussion

The sequential in vivo activation (p) of cortical MEK1/2 (p-MEK/total-MEK ratio) to p-ERK (p-ERK/total-ERK ratio) kinases was disrupted in mouse brain cortex after the acute regulation with various, and markedly different, cannabinoid CB₂ receptor ligands (i.e., direct agonist JWH133: modest decrease of MEK with ERK upregulation; endogenous agonist acting on overexpressed receptors: MEK downregulation with ERK upregulation; and direct inverse agonist/antagonist AM630: MEK upregulation with ERK unaltered) (see Figure 2). In the endocannabinoid system, these paradoxical findings for the CB₂ receptor contrast with the well-known canonical activation from MEK to ERK signaling after the stimulation of CB₁ receptors with the agonist WIN55212-2 (both kinases were concomitantly upregulated in rodent brain cortex [7, 8]).

Display full size

Figure 2. 

Schematic diagram depicting the disruption of the sequential regulation of MAP kinases p-MEK1/2 (ERK, MAP kinase) and p-ERK1/2 (phosphorylation and dephosphorylation processes) by CB₂ receptor ligands (full agonist, JWH133; inverse agonist, AM630); and CB₂ receptor overexpression (CB₂xP) in mouse brain cortex. MKPs: a family of specific MAP phosphatases; SL-327: a selective MEK1/2 inhibitor; ↓: down-regulatory effects; ↑: up-regulatory effects; =: lack-regulatory effects

In the paradoxical context of MEK-ERK physiological regulation, it is important to remark that the p-MEK1/2 kinases (i.e., MAP-ERK1 and 2 kinases) are equally competent activators of their unique substrates p-ERK1 and 2, and consequently, the cellular functions of p-ERK1/2 are under the tight control of their upstream activators p-MEK1/2 [1, 5]. In fact, the acute treatment with the selective MEK1/2 inhibitor SL-327 blocked the concomitant p-ERK activity in the mouse brain [7] (Figure 2). In the endocannabinoid system, and as mentioned above, the canonical activation of p-MEK to p-ERK signaling has been well demonstrated after stimulation of CB₁ receptors with selective full agonists (e.g., WIN55212-2) in rat and mouse brain cortex [7, 8]. Therefore, it was shocking to observe that treatment of mice with a selective CB₂ receptor agonist (JWH133) resulted in disruption of p-MEK (decreased) to p-ERK (increased) signaling in the brain cortex (Figure 1A and Figure 2). This unexpected result prompted the in vivo assessment of other CB₂ ligands for further possible divergent regulation of MEK-ERK in the brain.

Therefore, the major findings of this report demonstrate that the sequential phosphorylation of cortical MEK (ratio p-MEK1/2/total-MEK1/2) to p-ERK (ratio p-ERK1/2/total-ERK1/2) kinases are disrupted not only by the full CB₂ receptor agonist JWH133 (i.e., decreased MEK with increased ERK; Figure 1A and Figure 2) but also by acute treatments with the CB₂ inverse agonist/antagonist AM630, which as expected showed the opposite results to those induced by the cannabinoid agonist (i.e., increased MEK with unaltered ERK) (Figure 1C and Figure 2). In fact, an inverse agonist [10] is a ligand (e.g., AM630) that exerts the opposite pharmacological effects to a full receptor agonist (e.g., JWH133; see other relevant details in [13]). In fact, inverse agonism is better observed in those receptors that display high constitutive activity [10, 11, 13].

In this later context, and remarkably, the significant overexpression of cortical CB₂ receptor density (by 38%) in CB₂xP transgenic mice (see the second figure in [13]), which clearly favored the expression of CB₂ receptor constitutive activity with a change in receptor function (e.g., reduced basal expression of the pro-apoptotic MAP-kinase JNK1/2; see [13]), also resulted in a striking disruption of MEK to ERK parallel activity in brain cortex (i.e., marked decrease of MEK with marked increase of ERK) (Figure 1B and Figure 2). This result is in line with the reported agonist effect of JWH133 in normal CD1 mice (with a low level of CB₂ receptor expression), and suggests that after CB₂ receptor overexpression (38%) the sustained basal effects induced by the cannabinoid endogenous agonists (anandamide and/or 2-arachidonoylglycerol) on MEK-ERK would be more evident as clearly it was observed (Figure 1B and Figure 2). In this context, the biochemical and functional effects of a marked receptor overexpression (up to 195 times) were first reported for the cardiac β2-adrenoceptor, which resulted in a greater myocardial function in mice (see the pioneer study by Milano et al. [15]). In the present study and in the opposite side of the above reasoning, it could have been expected that a marked decrease of cortical p-MEK activity in mice (Figure 1B and Figure 2) would be translated into a substantial reduction in the activation of its exclusive substrate p-ERK (see [3–5]). However, this anticipated theoretical result was not observed (Figure 2) further suggesting a disruption of MEK-ERK signaling in the mouse brain cortex.

The reported in vivo disruption of the sequential and parallel basal activation of p-MEK to p-ERK by CB₂ receptor ligands had been observed for other Gi/o-coupled receptor and other experimental paradigms in rats and mice, including the treatment with low doses of the opioid receptor agonist fentanyl [16] as well as after treatments with structurally diverse hypnotic/anesthetic agents, which function through the allosteric activation of inhibitory gamma-aminobutyric acid subtype A (GABAA) receptors (see [14]). Interestingly, the acute treatment of mice with the irreversible (alkylating) receptor agent BU99006 (20 mg/kg, i.p., 1 h) also resulted in disruption of the expected parallel activity of MEK1/2 (p-MEK/total-MEK ratio decreased by 32%) to ERK1/2 (p-ERK/total-ERK ratio unaltered) signaling in the hippocampus [17]. In these previous studies, various molecular mechanisms were investigated in order to explain the disruption of brain MEK-ERK sequential regulation [for example, the roles of inactivated p-Thr286 MEK1 and specific MAP kinase phosphatases such as MAP kinase-1 (MPK-1), MPK-2, and MPK-3] [14–16], but the complex underlying in vivo mechanisms were not fully disclosed.

It is therefore unlikely that similar contra regulatory mechanisms could explain the remarkable disruption from MEK to ERK phosphorylation induced by the different cannabinoid CB₂ ligands in the mouse brain cortex (Figure 1). It is known that the stimulation of cannabinoids CB₂ receptors can engage multiple waves of different signaling pathways [12, 13], but at present the molecular signals underpinning the reported in vivo MEK-ERK dysregulation in the brain remain unknown. Moreover, some in vitro studies have reported that p-ERK1/2 signaling in rat-1 fibroblasts can be activated (with very high concentrations of peroxynitrites) via some MEK-independent pathway [18], but such a possibility contradicts the accepted postulates (mainly for in vivo experiments; e.g., see [7, 8] for specific details) of the canonical activation of MEK-ERK signaling (i.e., p-ERK1/2 are the sole cytoplasmic substrates of p-MEK1/2 kinases [3, 4], which have been validated by quantitative phosphoproteomic analysis [19]).

Further studies are needed to unravel the molecular mechanisms underpinning the striking disruption of MEK to ERK parallel regulation in the cannabinoid CB₂-Gi/o receptor signaling complex in the brain. Thus, a better understanding of the complexity of MEK-ERK in vivo signaling (e.g., assessing the effects of more efficacious synthetic CB₂ agonists and inverse agonists modulating neuronal death in psychiatric and neurodegenerative syndromes) would clarify the postulated roles of brain CB₂ receptors in various CNS diseases and their treatments with specific drugs [20, 21].

In this context, it has recently been suggested that the chronic use (during many years) of common medications by the elderly (e.g., antihypertensive, antidepressant, hypoglycemic, and hypocholesterolemic drugs [22]), and/or the abuse of alcohol (binge drinking and alcoholism) and/or hypnotic agents among older adults (drugs which also induce disruptions of the basal activation of p-MEK to p-ERK [14] and other references therein), might have a negative impact on neuronal survival accelerating the rate of progression of apoptotic death [23] in neuropsychiatric disorders such as Alzheimer and Parkinson syndromes, mainly because the very limited capacity of adult neurons to proliferate or be replaced.