Background
Acetaminophen (N-acetyl-para-aminophenol or paracetamol) was introduced in the market more than a century ago. This compound is one of the most common prescribed and over-the-counter (OTC) drugs in the world, although its mechanism of action is not fully understood. It has been shown that acetaminophen has good analgesic and antipyretic properties, but a weak anti-inflammatory activity, inhibiting the prostaglandin synthesis in the central nervous system (CNS) [
1], but not peripherally [
2].
Acetaminophen undergoes a deacetylation to
p-aminophenol not only in the liver but also in the CNS [
3]. In the CNS, p-aminophenol is conjugated with arachidonic acid by the fatty acid amide hydrolase (FAAH) to produce
N-arachidonoylphenolamine (AM404) [
3‐
5]. It has been suggested that AM404 may be responsible for the analgesic mechanism of paracetamol [
3,
6,
7].
The pharmacokinetics of AM404 formation in the CNS after acetaminophen administration has been studied by two different groups, Högestätt et al. [
3] and Murasamatsu et al. [
7]. Högestätt and collaborators have shown after 20 min of intraperitoneal injections of acetaminophen (30, 100, and 300 mg/kg) or p-aminophenol (10, 30, and 100 mg/kg), these compounds were converted to AM404 at the doses of 0.14, 1.6, and 10.3 pmol/g and 3.2, 44, and 667 pmol/g, respectively [
3]. Murasamatsu et al. (2016) demonstrated the conversion of acetaminophen in AM404 in rats treated orally with acetaminophen (20 mg/kg), and the peak of AM404 concentration was 150 pg/g at the half-life value of 0.3 h.
AM404 is a potent agonist of the transient receptor potential vanilloid type 1 (TRPV1) [
8], a low-affinity ligand of the cannabinoid receptor type 1 (CB1) [
3,
9], and an anandamide membrane transporter (AMT) blocker [
10,
11]. Furthermore, it has been shown that AM404 induces hypothermia [
12,
13] and analgesia in animal models [
14‐
16].
Some studies have demonstrated the effect of AM404 in modulating inflammation and oxidative stress. Its effects on reducing oxidative stress have been associated with the presence of a phenolic group in its structure [
17,
18]. AM404 decreased the production of the cytokines interleukin (IL)-1β and IL-6 and increased in circulating tumor necrosis factor (TNF)-α levels in a murine model of inflammation induced by LPS injection [
19]. In a rat model of neuropathic pain, AM404 prevented the overproduction of nitric oxide (NO) and TNF-α and increased IL-10 production [
14]. Furthermore, this acetaminophen metabolite inhibited the prostaglandin (PG) E
2 formation and the activity of isolated cyclooxygenase (COX)-1 and COX-2 enzymes ex vitro, and in RAW264.7 macrophages [
3]. In human T cells, AM404 is a potent inhibitor of T cell receptor (TCR)-mediated T cell activation and specifically inhibited both IL-2 and TNF-α gene transcription and TNF-α protein synthesis in CD3/CD28-stimulated Jurkat T cells in a FAAH independent way regulating the activation of the transcription factors NF-κB, NFAT, and AP-1 [
20]. In an animal model, it has been shown to inhibit both isoforms of prostaglandin endoperoxide synthase/cyclooxygenases (PGHS/COX-1 and -2) [
21].
Although there are few studies suggesting that AM404 modulates inflammation, the effects of AM404 on the production of members of the arachidonic acid pathway in activated microglia and brain tissues have not been elucidated in detail, possible explaining some mechanism of actions of acetaminophen in the CNS. We therefore investigated the effects of AM404 on the production of eicosanoids induced by LPS in organotypic hippocampal slice cultures (OHSC) and activated primary microglia.
Methods
Reagents
AM404 was obtained from Tocris Biosciences (Ellisville, MO) and dissolved in dimethyl sulfoxide (DMSO - Merck KGaA, Darmstadt, Germany) to get a 50 mM stock solution. Further dilutions in DMSO were prepared immediately before the stimulation. AM251 (CB1 antagonist) and capsazepine (TRPV1 antagonist) (Tocris Biosciences) were dissolved in DMSO. LPS from Salmonella typhimurium (Sigma-Aldrich, Taufkirchen, Germany) was resuspended in Dulbecco’s Phosphate Buffered Saline (DPBS; Gibco® by Life Technologies, Germany) as 5 mg/mL stock and was used at a final concentration of 10 ng/mL in the microglial culture and 100 ng/mL in OHSC. Solvent concentration in the culture media was maintained at 0.1%.
Preparation of organotypic hippocampal slice cultures (OHSC)
OHSC of C57Bl/6 wild-type (WT) mice were prepared as previously described [
22]. In brief, 2 to 3 days old animals, under sterile conditions, were decapitated; the hippocampi were dissected and placed in a tissue chopper (McIlwain™) for the preparation of 350 μm slices. The slices were transferred to a 0.4 μm culture plate inserts (Millipore, PICM03050). Each insert containing 6 slices were placed in a six-well culture plate containing 1 mL of culture medium [0.5× minimum essential medium (MEM), 25% horse serum, 25% basal medium without glutamate (BME), 2 mM glutamax, and 0.35% glucose]. Then, the slices were incubated at 35 °C in humidified atmosphere with 5% CO
2 (Heracell 240i, Thermo Scientific). The culture medium was changed after the first day of preparation following every 2 days. After 1 week, the slices were pre-incubated with DMSO 0.1% or AM404 50 μM [
23] for 30 min and then stimulated with LPS for 24 h.
Depletion of microglia from slices culture
For the depletion of the microglia from the slice cultures, 1:10 of liposome-encapsulated clodronate (Lip-CL - Merck Chemicals GmbH, Darmstadt, Germany) solution (0.5 mg/mL) was used. After preparation of OHSC, as described above, the slices were incubated overnight (35 °C in humidified atmosphere with 5% CO
2 - Heracell 240i, Thermo Scientific) with culture medium containing Lip-CL. Then, slices were rinsed in DPBS at 35 °C and cultured in standard culture medium for another 6 days [
22]. This procedure reduces the microglia population to less than 5%, but it does not affect the astrocytes and neuron cells [
22,
24].
Immunohistochemistry
After 1 week of OHSC preparation (wild-type––WT or microglia depleted––Lip-CL), OHSC were washed with DPBS followed by 4% paraformaldehyde (PFA) incubation for 1 h. After fixation, the slices were washed with DPBS and incubated with 5% normal goat serum (NGS - Vector) in DPBS containing 0.3% Triton X-100 (DPBS+) for at least 2 h. Subsequently, the slices were incubated overnight with mouse anti-GFAP (1:1000, Cell signaling), anti-rabbit-Iba-1 (1:1000, Wako), and DAPI (1:1000, Sigma) in 1% NGS/DPBS+ at 4 °C. Then, slices were incubated with the secondary antibodies for 2 h at room temperature. Rabbit highly cross-adsorbed AlexaFluor 594, and mouse AlexaFluor 488 secondary antibody (Invitrogen, Carlsbad, CA, USA) was used to detect Iba-1 or GFAP, respectively. Slices were imaged in a Zeiss microscope (Zeiss, Oberkochen, Germany).
Primary rat and mouse microglial cell culture
Primary mixed glial cell cultures were prepared from cerebral cortices of 1-day neonatal Sprague-Dawley rats or C57Bl/6 mice or TPRV1
−/− knockout mice, as described previously [
25,
26]. Briefly, 7–13 forebrains were removed, homogenized, and filtered through a 70-μm cell strainer (BD biosciences, Heidelberg, Germany), under sterile conditions. After centrifugation (1000 rpm, 10 min), cells were collected and resuspended in Dulbecco’s Modified Eagle’s Medium (DMEM) containing 10% fetal calf serum (Biochrom AG, Berlin, Germany) and antibiotics (40 U/mL penicillin and 40 μg/mL streptomycin, both from PAA Laboratories, Linz, Austria) and cultured (5 × 10
5 cells/plate) on 10-cm cell culture dishes (Falcon, Heidelberg, Germany) in 10% CO
2 at 37 °C (Heracell 240i, Thermo Scientific). Floating microglia were harvested every week (2–7 weeks) and re-seeded into 75 cm
2 culture flask to give pure microglial cultures. On the next day, medium was changed to remove non-adherent cells, and after 1 h, the cells were stimulated for respective experiments.
Cytotoxicity assay
Cytotoxicity assay was performed using CellTox™ Green Cytotoxicity assay kit (Promega, Mannheim, Germany). Briefly, cells were cultured in 96-well plates at the density of 25 × 103 cells/well in DMEM medium containing 10% fetal calf serum (Biochrom AG, Berlin, Germany) and antibiotics (40 U/mL penicillin and 40 μg/mL streptomycin, both from PAA Laboratories, Linz, Austria). Cells were pre-treated with different concentrations of AM404 (0.1–10 μM) or DMSO 0.1% for 30 min. Thereafter, cells were incubated with or without LPS for the next 24 h. Ethanol (10% end conc., Sigma-Aldrich, Taufkirchen, Germany) was used as positive control to induce the cell death. After incubation (10% CO2 at 37 °C - Heracell 240i, Thermo Scientific), 100 μl of CellTox™ Green reagent were added in each well. The plate was mixed for 1 min and incubated for 15 min at room temperature, and the fluorescence was measured at 490 nmEx/530 nmEm using a Modulus™ II Microplate Multimode Reader (Turner BioSystems, USA).
The principle of the assay is to evaluate the alterations in the membrane integrity, using the cyanine dye. The dye binds in the dead-cell DNA and enhanced the fluorescent property, which is excluded from viable cells. The fluorescence intensity values obtained were normalized and presented as the percentage of untreated controls.
Determination of prostaglandin (PG) D2, PGE2, and 8-iso-prostaglandin F2α (8-iso-PGF2α) release by enzyme-linked immunosorbent assay (ELISA)
Microglial cells were pre-treated for 30 min with different concentrations of AM404 (0.1–10 μM) or DMSO 0.1%. Thereafter, LPS (10 ng/mL) was added for 24 h. Supernatants were harvested, and levels of PGE2 (Assay Designs Inc., Ann Arbor, MI, USA; distributed by Biotrend, Cologne, Germany), PGD2, and 8-iso-PGF2α (Cayman Chemicals, Ann Arbor, Michigan, USA) were measured by enzyme immunoassay (EIA) in the medium according to the manufacturer’s instructions. Standards from 7.8 to 500 pg/mL, 78 to 10,000 pg/mL, and 3.9 to 500 pg/mL were used for PGE2 (detection limit of 3.25 pg/mL), PGD2 (detection limit of 200 pg/mL) and 8-iso-PGF2α (detection limit of 2.7 pg/mL), respectively. Data was normalized to LPS and presented as percentage of change in PGs levels of at least three independent experiments.
Western blot analysis
For COX-2 and mPGES-1 immunoblotting, microglial cells were left untreated or treated with LPS (10 ng/mL) in the presence or absence of AM404 (0.1–10 μM) for 24 h. Cells were washed with cold phosphate buffered saline (PBS) and added lysis buffer (42 mM Tris-HCl, 1.3% sodium dodecyl sulfate, 6.5% glycerin, 100 μM sodium orthovanadate, and 2% phosphatase and protease inhibitors). Protein estimation was measured using the bicinchoninic acid method (BCA protein determination kit from Pierce, distributed by KFC Chemikalien, Munich, Germany) according to manufacturer’s instructions and bovine serum albumin (BSA, Sigma) used as a standard. For Western blotting, 20 μg of total protein from each sample were subjected to sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) under reducing conditions. Then, proteins were transferred onto a polyvinylidene fluoride (PVDF) membrane (Merck Millipore, Darmstadt, Germany) by semi-dry blotting. The membrane was blocked for 1 h at room temperature using Rotiblock (Roth, Karlsruhe, Germany) and incubated with the primary antibody overnight. Primary antibodies were goat anti-COX-2 (M19, 1:500, Santa Cruz Biotechnology, CA, USA), rabbit anti-mPGES-1 (1:200, Cayman Chemical Co., Ann Arbor, MI, USA), and rabbit anti-actin IgG (1:5000, Sigma, Saint Louis, MO, USA). AB-coupled proteins were detected with horseradish peroxidase-coupled rabbit anti-goat IgG (Santa Cruz, 1:100,000 dilution) or goat anti-rabbit IgG (Amersham, 1:25,000 dilution) using enhanced chemiluminescence Western blotting substrate (Biozym Scientific GmbH, Rockford, USA). The quantification of the Western blots was performed using ImageJ. All Western blot experiments were carried out at least three times.
Cyclooxygenase activity assay
To evaluate the effect of AM404 on COX enzymatic activity, an arachidonic acid assay was performed as previously described [
27]. There are currently two well identified cyclooxygenase isoforms, named COX-1 and COX-2. The first one is constitutively expressed in almost every cell and COX-2, although can also be constitutive in some cells, is induced by cytokines and endotoxins [
28,
29]. Under unstimulated conditions, primary microglial cells only express the COX-1 isoform [
30]. Briefly, to measure COX-1 activity, primary rat microglial cells were plated in 24-well cell culture plates, and after 24 h, the medium was removed and replaced with serum-free medium. AM404 (0.1–10 µM) or selective COX-1 inhibitors [acetylsalicylic acid (ASA, 50–100 μM), irreversible inhibitor, or SC560 (0.1–1 μM) reversible inhibitor] were added, and left for 15 min. After that, 15 µM of arachidonic acid were supplemented for another 15 min. Supernatants were then collected and used for the determination of PGE
2.
We also investigated the effects of AM404 on microglial total COX (COX-1 + COX-2) enzymatic activity, which mainly determined COX-2 mediated PGE2 production. Since we also have a slight contribution of COX-1 activity in this assay, we titled as COX-1 + COX-2 activity.
The COX-1 + COX-2 activity assay was conducted exactly as mentioned before with pre-incubation with LPS (10 ng/mL) for 24 h to induce COX-2 synthesis and using diclofenac sodium (preferential COX-2 inhibitor, 1–10 μM).
Furthermore, to understand if the effects of AM404 in COX-1 and COX-1+COX-2 activities produce a reversible or irreversible inhibition, the assays were evaluated with some modifications. For COX-1 activity, the drugs were incubated for 60 min, after the wells were washed three times with PBS 37 °C and incubated for 3 h with serum-free medium. Then, medium was changed and added 15 µM arachidonic acid for 15 min. For total COX activity, primary rat microglial cells were first pre-incubated with LPS (10 ng/mL) for 24 h and then follow the same procedure describe for COX-1. Finally, the supernatants were collected for determination of PGE2.
Data analysis
Results were converted into percentage values of LPS and presented as mean ± SEM. Data was analyzed using one-way analysis of variance (ANOVA) followed by Newman-Keuls post-test. The level of statistical significance was set as *p < 0.05, **p < 0.01, and ***p < 0.001. Graph Pad Prism (Graph Pad Software, San Diego, CA) was used for performing all statistical analysis.
Discussion
In the present study, we demonstrated that the acetaminophen metabolite, AM404, reduces inflammatory mediators of the arachidonic acid cascade in OHSC and primary microglial cells. We show here that AM404 decreased the release of PGD2, PGE2, and 8-iso-PGF2α, by a mechanism independent of TRPV1 or CB1 receptors, but by reversibly inhibiting COX-1 and COX-2 activity, as well as by slightly reducing the expression levels of COX-2 protein in microglia.
Various studies have been performed to elucidate the mechanism of action of acetaminophen since it is one of the most prescribed and consumed drugs in the world for mild to moderate pain relieve and fever reduction, and its intracellular pathways are not fully understood. Acetaminophen produces a weak inhibition of COX, and it has been proposed that acetaminophen acts on COX-3, a brain-specific isoform of COX and slice variant of constitutive enzyme COX-1 [
32,
33]. Indeed, many studies have demonstrated its inhibitory effect on PGE
2 production in the brain [
1] and in primary microglial culture [
2,
34]. PGE
2 is an important subproduct of the enzymatic action of COX and a mediator of fever where its increase occurs in response to systemic pyrogen agents produced by infectious pathogens [
35‐
38].
Högestätt et al. [
3] and Murasamatsu et al. [
7] have demonstrated that milligrams of paracetamol are converted in picograms of AM404 into the CNS of rodents, which could induce different pharmacological effects, including the effects on COX expression and activity. However, the correlation between in vitro and in vivo studies is complex due to different pharmacokinetic and micro-environment parameters to be considered. In the meantime, it has been suggested that AM404 is the active metabolite of acetaminophen and is responsible for its analgesic activity [
3,
6,
7]. The involvement of AM404 in analgesia has been well established in animal models of nociceptive and neuropathic pain [
14‐
16,
39]. In addition, the effect of AM404 in modulating peripheral inflammation [
14,
19,
21] and in immune cell culture [
3,
20] has been described. However, the direct effect of AM404 in neuroinflammation, especially in microglia, is still poorly understood.
OHSC is a well-established brain tissue model maintaining the main architecture of the cells, and its complex system provides a good tool to understand cellular and molecular processes of the brain in vitro [
40,
41]. Our results in OHSC demonstrated that the AM404 inhibited the release of PGE
2 induced by LPS stimulation, and by depleting the microglial cells from OHSC, the levels of PGE
2 after stimulation with LPS were completely abolished. As known, microglial cells are the resident macrophages of the CNS and the most important source of PGE
2 in neuroinflammation. It has been demonstrated before that in primary microglial cells stimulated with LPS, PGE
2 and COX-2 expression are strongly increased [
42‐
44].
Thus, to elucidate the mechanisms by which AM404 affects these cells, we continued our further experiments with primary microglial cells (mouse and rat) treated with LPS. Studies in primary cell cultures allowed us to better understanding of the mechanism of pure microglia as a result of distinct stimulations. It has been reported that the process of microglia isolation may lead to their activation [
45,
46]. However, we do not observe an activation status in respect of inflammatory parameters. After LPS stimulation, we observe comparable data in cultivated microglia mono-cultures and in organotypic slice culture models, where LPS-induced PGE
2 release, and therefore believe that our results are robust in both models, although the morphological and biochemical status of microglia may depend on the cells that are also present in the cultured [
46‐
50]. The pre-treatment with AM404 potently prevented the increase on PGE
2 release stimulated by LPS from both species. This effect of AM404 on downregulating PGE
2 release provides us an important contribution to explain the mechanism by which paracetamol might control fever. The involvement of AM404 in inhibition of LPS-induced PGE
2 formation in RAW264.7 macrophages was also described by Högestätt and colleagues [
3].
AM404 modulates the endocannabinoid system by acting via TRPV1 and CB1 receptors, and it increased the viability of anandamide in the medium by blocking AMT [
10,
11]. The involvement of cannabinoids in inflammation has been shown in different models and demonstrated some evidences for anti-inflammatory effects [
51,
52]. Furthermore, many studies have described the participation of cannabinoids on immune modulation [
53‐
56]. Immune cells express the cannabinoid receptors type 1 and type 2, albeit CB2 appears to be the predominant isoform [
57,
58]. In microglia, the expression and the amount of CB receptors may change depending on the type of neuropathology. However, it has been reported that rat and mouse primary microglia cultures express both CB1 and CB2 receptor mRNA and protein [
52]. Moreover, microglia can also express TRPV1 that mediate some endocannabinoid actions [
59,
60]. However, the involvement of TRPV1 and CB1 receptors in AM404 effects is still not completely elucidated since the present data are controversial. Roche et al. [
19] have shown that the effect of AM404 on the increase of circulating TNF-α after LPS injection was blocked by SB366791 (TRPV1 antagonist) and AM251 (CB1 antagonist) and the decrease on IL-1ß was attenuated only by AM251, but neither of these antagonists altered the effect of AM404 on the decrease of IL-6. In addition, in a model of cerebral ischemia in gerbils, AM404 reduced neuronal damage, and this effect was reversed by AM251, but not by capsazepine [
61]. In the present study, we evaluated the involvement of TRPV1 and CB1 receptors in the modulation of PGE
2 release by AM404, and our findings showed that the mechanisms underlying the anti-inflammatory effect of AM404 in CNS cells did not involve TRPV1 and CB1 receptors, additionally supported by using TRPV1 knockout mice.
Free radicals also contribute to the inflammatory process by metabolizing arachidonic acid to isoprostanes. The involvement of AM404 on the decrease of oxidative stress has been proposed by the presence of a phenolic group in its structure [
17,
18], and Costa et al. [
14] proved in a model of neuropathic pain that AM404 prevented the overproduction of nitic oxide (NO). In addition, García-Arencibia and collaborators (2007) have suggested the antioxidant effect of AM404 as one of the neuroprotective mechanism of AM404 in a Parkinson’s disease [
62] model. In accordance with the literature, we showed that AM404 prevented the formation of reactive oxygen species in primary microglial cells.
In order to understand the mechanism by which prostaglandins are reduced by AM404, we tested the hypothesis that AM404 inhibits COX activity as described in a monocyte cell line stimulated with LPS (1 μg/mL) [
3]. Indeed, we demonstrated that AM404 inhibited COX-1 and COX-2 activities. Furthermore, the inhibition of COX seems to be reversible. Additionally, this compound slightly reduced COX-2 protein levels induced by LPS in microglia. Thus, different mechanisms may participate in the reduction of prostaglandin levels mediated by AM404.