Acetaminophen inhibits diacylglycerol lipase synthesis of 2-arachidonoyl glycerol: Implications for nociception

Abstract

Acetaminophen (paracetamol) is a common analgesic, but its mechanism of action remains unknown. Despite causing around 500 deaths annually in the US, safer alternatives have not been developed. Because endocannabinoids may have a role in acetaminophen action, we examine interactions between the two. We report that acetaminophen inhibits the activity of diacylglycerol lipase α (DAGLα), but not DAGLβ, decreasing the production of the endocannabinoid 2-arachidonoyl glycerol. This gives rise to the counterintuitive hypothesis that decreasing endocannabinoid production by DAGLα inhibition may be antinociceptive in certain settings. Supporting this hypothesis, we find that diacylglycerol lipase (DAGL) inhibition by RHC80267 is antinociceptive in wild-type but not CB1 knockout mice in the hot-plate test. We propose (1) that activation of DAGLα may exacerbate some forms of nociception and (2) a mechanism for the antinociceptive actions of acetaminophen, whereby acetaminophen inhibits a DAGLα/CB1-based circuit that plays a permissive role in at least one form of nociception.

Keywords: 2-arachidonoyl glycerol; acetaminophen; antinociception; cannabinoid receptor; diacylglycerol lipase; paracetamol.

Graphical abstract

Figure 1

Acetaminophen inhibits endogenous cannabinoid neuroplasticity in autaptic hippocampal neurons (A) Acetaminophen does not alter excitatory neurotransmission as measured by evoked EPSC charge. (B) Acetaminophen inhibits endocannabinoid-mediated depolarization-induced suppression of excitation (DSE) in a concentration-dependent manner. (C) Sample time courses of DSE in response to a 3 s depolarization before and during 100 μM acetaminophen. Inset shows representative EPSC traces for control (left) and 100 μM acetaminophen (right) after a 3 s depolarization (scale bars: 1 nA and 5 min). (D) Responses to the GABA-B agonist baclofen (25 μM) are not altered by 100 μM acetaminophen pretreatment. (E) Sample time course for baclofen inhibition of EPSCs in the presence of acetaminophen (100 μM). (F) AM404 (500 nM) does not affect the magnitude of DSE. n = 4–6. Data are represented as mean ± SEM.

Figure 2

Acetaminophen does not act presynaptically to inhibit CB1 signaling (A) EPSCs are inhibited by ∼40% by 2-AG (1 μM). This inhibition is not reversed by treatment with 100 μM acetaminophen. (B) Sample time course shows that 2-AG inhibition of EPSCs is not reversed by acetaminophen treatment. Right shows representative EPSC traces for control (A), 2-AG (B), and 2-AG + 100 μM acetaminophen (C) (scale bars: 1 nA and 10 ms). n = 7. Data are represented as mean ± SEM.

Figure 3

Both acetaminophen and DO34 reduce 2-AG production in HEK-DAGLα cells Levels of 2-AG (A), 2-OG (B), 2-LG (C), AEA (D), OEA (E), PEA (F), and LEA (G) are shown for HEK-DAGLα cells treated with S-AG (25 μM) + vehicle (VEH), S-AG (25 μM) + acetaminophen (APAP, 100 μM), or S-AG (25 μM) + DO34 (DO34, 1 μM). ∗p < 0.05, ∗∗p < 0.01 by one-way ANOVA. n = 6. Data are represented as mean ± SEM.

Figure 4

Acetaminophen reduces DAGLα activity (A) A sample time course of EnzChek lipase activity assay in lysates from DAGLα-transfected HEK cells shows that acetaminophen (30 μM) reduces net lipase activity. (B) Acetaminophen (30 μM) does not reduce activity in control HEK cell lysates. (C) In DAGLβ-transfected cells, acetaminophen does not reduce activity. (D) Summary graph showing responses for 30 μM acetaminophen in WT and DAGLα- or DAGLβ-transfected HEK293 cells. ∗p < 0.05 by unpaired t test, n = 4–5. Data are represented as mean ± SEM.

Figure 5

Acetaminophen inhibits muscarinic endocannabinoid production in HEK293 cells co-expressing an eCB sensor and DAGLα (A) Summary of experiments showing muscarinic agonist oxotremorine-M (oxo-M, 10 μM) responses relative to 2-AG in same cell after vehicle, acetaminophen (APAP, 50 μM), or DO34 (300 nM). (B) Sample time course showing that HEK293 cells expressing only eCB3.0 do not respond to 10 μM oxo-M but robustly respond to 2-AG (5 μM). (C) Summary of experiments showing that APAP (50 μM) and DO-34 (300 nM) each reduce the effect of oxo-M in HEK293 cells co-expressing eCB3.0 and DAGLα. (D) Sample time course in HEK293 cells co-expressing eCB3.0 and DAGLα shows responses to oxo-M, 10 μM, and subsequent 2-AG. Treatment with APAP (50 μM, 10 min) reduces this response as a fraction of 2-AG response in the same experiment. Time courses are normalized to 2-AG. (E) Sample time course showing the effect of DO34 (300 nM) treatment in HEK293 cells co-expressing eCB3.0 and DAGLα. ∗∗∗p < 0.005, ∗∗∗∗p < 0.0001 by one-way ANOVA with Dunnett’s post hoc test vs. vehicle. n = 12, 7, 8. Data are represented as mean ± SEM.

Figure 6

Acetaminophen and the DAGL inhibitor RHC-80267 are antinociceptive in wild-type but not CB1 knockout mice (A) Acetaminophen (300 mg/kg; i.p.) increased paw withdrawal latency in the hot-plate assay relative to baseline in WT but not CB1 KO male mice. (B) Same experiment as in (A) for female mice. (C) DAGLα/β inhibitor, RHC-80267 (20 mg/kg i.p.), prolonged the paw withdrawal latency in WT but not CB1 KO male mice. (D) Same experiment as in (C) for female mice. ∗p < 0.05, ∗∗∗p < 0.001 by paired t test vs. pre-drug baseline, n = 6 mice per group. Data are represented as mean ± SEM.

Figure 7

Proposed role for DAGLα and 2-AG in nociception A schematic depiction of current and proposed acetaminophen (APAP) actions.