Chemical probes to potently and selectively inhibit endocannabinoid cellular reuptake

Edited by Benjamin F. Cravatt, The Scripps Research Institute, La Jolla, CA, and approved May 10, 2017 (received for review March 14, 2017)
June 5, 2017
114 (25) E5006-E5015


Suitable chemical tools have been instrumental in the discovery and characterization of the endocannabinoid system. However, the lack of potent and selective inhibitors for endocannabinoid transport has prevented the molecular characterization of this process. Current uptake inhibitors are poorly bioavailable to the central nervous system (CNS) and weakly selective because they also inhibit fatty acid amide hydrolase (FAAH), the major anandamide-degrading enzyme. Few studies have addressed the uptake inhibition of 2-arachidonoyl glycerol (2-AG), which is the major endocannabinoid. Here, we report a highly potent and selective endocannabinoid reuptake inhibitor. Our data indicate that endocannabinoid transport across the membrane can be targeted, leading to general antiinflammatory and anxiolytic effects in mice.


The extracellular effects of the endocannabinoids anandamide and 2-arachidonoyl glycerol are terminated by enzymatic hydrolysis after crossing cellular membranes by facilitated diffusion. The lack of potent and selective inhibitors for endocannabinoid transport has prevented the molecular characterization of this process, thus hindering its biochemical investigation and pharmacological exploitation. Here, we report the design, chemical synthesis, and biological profiling of natural product-derived N-substituted 2,4-dodecadienamides as a selective endocannabinoid uptake inhibitor. The highly potent (IC50 = 10 nM) inhibitor N-(3,4-dimethoxyphenyl)ethyl amide (WOBE437) exerted pronounced cannabinoid receptor-dependent anxiolytic, antiinflammatory, and analgesic effects in mice by increasing endocannabinoid levels. A tailored WOBE437-derived diazirine-containing photoaffinity probe (RX-055) irreversibly blocked membrane transport of both endocannabinoids, providing mechanistic insights into this complex process. Moreover, RX-055 exerted site-specific anxiolytic effects on in situ photoactivation in the brain. This study describes suitable inhibitors to target endocannabinoid membrane trafficking and uncovers an alternative endocannabinoid pharmacology.
The endocannabinoid system (ECS) is a pan-organ lipid signaling network that modulates numerous biological processes, including neurotransmission and immune function (1, 2). The major endogenous agonists [i.e., endocannabinoids (ECs)] for cannabinoid receptors CB1 and CB2 are the arachidonic acid (AA)-derived lipids 2-arachidonoyl glycerol (2-AG) and N-arachidonoylethanolamine [anandamide (AEA)]. Altered EC signaling in the brain has been implicated in nociception (3), learning and memory (4), anxiety (5), and depression (6). The indirect modulation of EC levels may lead to fewer side effects than the direct activation of CB1 receptors in terms of neurotransmission, metabolism, and immunomodulation (7).
CB1 receptor agonists are intrinsically associated with strong central side effects that are far less pronounced for increasing EC levels upon blockage of the main EC hydrolytic enzymes fatty acid amide hydrolase (FAAH) and monoacylglycerol lipase (MAGL). In addition to general antiinflammatory and analgesic effects, the modulation of EC tissue concentrations is a promising therapeutic approach to treat diseases related to the central nervous system (CNS) (8, 9). Pharmacological strategies to treat neuropsychiatric disorders currently focus on the inhibition of EC degradation (10). FAAH and MAGL inhibitors such as URB597 (11) and JZL184 (12), respectively, have been instrumental to elucidate the role of AEA and 2-AG in rodent models of anxiety and depression (6, 1214). Although AEA and 2-AG have different intracellular fates, they may share a common mechanism of membrane trafficking that is selective for ECs over arachidonate and other N-acylethanolamines (NAEs) (1519). However, although suitable inhibitors are available for most targets within the ECS (20), the existing AEA uptake inhibitors lack potency and show poor selectivity over the other components of the ECS, in particular FAAH (21, 22).
Given the lack of appropriate inhibitors, it is not unexpected that the process of EC cellular uptake has remained largely uncharacterized at the biochemical level and, therefore, is also controversially discussed (21). Here, building on previous work on N-alkyl-2,4-dodecadienamides from Echinacea purpurea (L.) Moench, which have been shown to interact with the ECS (22, 23), a series of derivatives and analogs of these natural unsaturated fatty acid amides were synthesized, and their effects on EC transport were investigated. This work has resulted in the identification of (2E,4E)-N-[2-(3,4-dimethoxyphenyl)ethyl]dodeca-2,4-dienamide (WOBE437; 1) as a highly potent and selective EC uptake inhibitor, which was extensively profiled. In addition, we have designed and synthesized the WOBE437-derived photoaffinity probe RX-055 (2) as a potent and irreversible EC uptake inhibitor, which has enabled unambiguous insights into the uptake process.


Discovery of WOBE437 as a Highly Potent AEA Uptake Inhibitor.

Based on the natural product (2E,4E)-N-isobutylamidedodeca-2,4-dienamide (3) (SI Appendix, Fig. S1 and Table S1) from the medicinal plant E. purpurea as a starting point, we synthesized a library of 634 analogs and derivatives with varying alkyl chain lengths and structure of the head group moiety (SI Appendix, Fig. S1 and SI Experimental Procedures). All new derivatives were tested for AEA uptake inhibition in U937 cells by using a screening assay (SI Appendix, SI Experimental Procedures). From a total of 348 analogs, we identified the dodeca-2E,4E-dienoyl N-alkylamide scaffold as the most promising framework for the development of EC transport inhibitors being highly selective over FAAH (>100-fold selectivity; Fig. 1 A and B). For instance, N-(pyridin-3-yl)ethyl, N-(2-methoxyphenyl)ethyl, and N-(2,3-dihydro-1,4-benzodioxin-6-yl)ethyl dodeca-2E,4E-dienamide (14, 20, and 25, respectively) exhibited IC50 values in the nanomolar range for AEA uptake and 60- to 105-fold selectivity over AEA hydrolysis (Fig. 1 A and B and SI Appendix, Table S1). The ethyl linker connecting the amide group with the aromatic system provided the optimal length for effective uptake inhibition, as exemplified by the comparison between the 14, pyridin-3-ylmethyl (15), and pyridin-3-yl (16) head groups, which led to IC50 values of 101; 2,691; and 1,242 nM, respectively (Fig. 1A and SI Appendix, Table S1). The most potent compounds were those with an N-phenethyl head group, which showed IC50 values in the low nanomolar range (SI Appendix, Table S1). Compared with the unsubstituted parent compound 8 (IC50 = 1,142 nM), the presence of a single methoxy group at different positions of the aryl system (20, 21, and 22) was associated with approximately fourfold increased potency (IC50 = 198–271 nM) (Fig. 1 A and B and SI Appendix, Table S1). Potency could be increased dramatically by 3,4-dimethoxylation, providing the highly potent and selective inhibitor 1 (WOBE437) with an IC50 value of 10 ± 8 nM for AEA uptake inhibition (using 100 nM total AEA) and an outstanding 1,000-fold selectivity over FAAH (Fig. 1 A and B and SI Appendix, Table S1). The structurally related benzodioxole (24) and dihydrobenzodioxine (25) analogs were ∼13 times less potent (Fig. 1 A and B and SI Appendix, Fig. S2). Modifications in the acyl part of WOBE437 (chain length and degree of unsaturation) led to less potent analogs (SI Appendix, Fig. S3). The similarity between the head groups in WOBE437 and the minor EC N-arachidonoyl dopamine (24) prompted us to also prepare (2E,4E)-dodecadienoyl dopamine (32) and the AA-based analog of WOBE437 (i.e., 34). Interestingly, 32 was ∼100-fold less potent than WOBE437, whereas 34 retained significant potency and was only sixfold less potent (SI Appendix, Table S1). Similarly, the (2E,4E)-dodecadienoyl ethanolamine (33) generated by linking the acyl chain of WOBE437 with the head group of AEA significantly lost potency (IC50 = 3.47 μM) (SI Appendix, Table S1).
Fig. 1.
Design and in vitro pharmacological characterization of the highly potent and selective EC transport inhibitor WOBE437. (A) Chemical structures of a selection of dodeca-2E,4E-dienoyl N-alkylamides and IC50 values for AEA uptake. WOBE437 is highlighted in red. (B) Correlation of IC50 values for AEA uptake and hydrolysis inhibition of the most potent N-alkylamides and reference compounds (FAAH inhibitors are in green; AEA uptake inhibitors are in blue). No significant correlation between these two processes (indicated with the gray dotted line) was observed [Pearson’s ρ = 0.258, P = 0.214, not significant (α = 0.05)]. WOBE437 (red) shows exceptional potency for AEA uptake inhibition and selectivity over FAAH inhibition. (C) Inhibition of AEA uptake by WOBE437 in Neuro2a cells (after 10 min) showing normalized data. (D) Inhibition of AEA and 2-AG uptake induced by WOBE437 in U937 cells (after 5 min) showing normalized data. Time course of AEA uptake in Neuro2a cells (E; 100 nM WOBE437 vs. vehicle) and primary rat cortical neurons (F; 1 μM WOBE437 vs. vehicle). (G) Time course of 2-AG uptake in Neuro2a cells in the presence of WOBE437 (5 μM) or vehicle. (H) WOBE437 does not inhibit any of the major serine hydrolases in membranes obtained from mouse brain (ABPP experiment). URB597, JZL184, JZL195, and WWL70 were used as positive controls for FAAH, MAGL, dual FAAH/MAGL, and ABHD6 inhibition, respectively. Data show mean values ± SD from at least three independent experiments, each performed in triplicate. For D, data show mean values ± SEM from 10 independent experiments, each performed in triplicate. Statistical significance was calculated with two-tailed unpaired t test. *P < 0.05; **P < 0.01 vs. vehicle. CTRL, control.

Biochemical and Pharmacological Profiling of WOBE437 Shows Functional Specificity.

In addition to their effects on AEA uptake, the inhibition of FAAH was thoroughly investigated for all nanomolar inhibitors (SI Appendix, Table S1). Several N-alkylamides exhibited dramatically improved selectivity for AEA uptake reduction over FAAH inhibition compared with any previously described nonselective and cell-permeable AEA uptake inhibitors (Fig. 1B and SI Appendix, Table S2). The lack of biologically significant FAAH inhibition by WOBE437 was established in different assay systems by using human recombinant enzyme, cell, and brain homogenates. The compound did not inhibit AEA hydrolysis in all biological matrices investigated at pharmacologically relevant concentrations (IC50 ≥ 10 μM) (SI Appendix, Fig. S4 A and B). Conversely, WOBE437 showed low nanomolar potency for the inhibition of AEA uptake in different cell lines and assay formats, including the FAAH-deficient HMC-1 human mast cells (25) (IC50 = 137 ± 31 nM, Hill slope = −0.934; SI Appendix, Fig. S5A and Table S3), Neuro2a mouse neuroblastoma cells (IC50 = 55 ± 18 nM, Hill slope = −0.705) (Fig. 1C), and U937 cells (IC50 = 10 ± 8 nM, Hill slope = −0.715) (Fig. 1D). Independent time-course experiments showed that in Neuro2a cells, WOBE437 (100 nM) inhibited the overall AEA uptake by ∼35% (nonnormalized data) after 2–10 min of incubation, reducing the effect to 20% inhibition after 15–20 min of incubation (Fig. 1E). Importantly, in a comparable assay, WOBE437 (1 μM) also inhibited AEA uptake in primary rat cortical neurons by 50% after 2–7 min of incubation with a reduced inhibition (20–35%) after 10–20 min of incubation (Fig. 1F). We and others have suggested that AEA and 2-AG compete for cellular uptake despite their differing intracellular fates (15, 17). In line with this concept, WOBE437 inhibited 2-AG uptake (using 1 µM total 2-AG) in U937 cells with an IC50 of 283 ± 121 nM (Hill slope = −0.973; Fig. 1D). In Neuro2a cells, incubation with WOBE437 (5 µM) at 2 and 5 min reduced overall 2-AG uptake by 40% (Fig. 1G). Interestingly, the α/β-hydrolase domain-6 (ABHD6) inhibitor WWL70 prevented 2-AG uptake and hydrolysis occurring at later time points, unlike JZL184, which specifically inhibited hydrolysis (SI Appendix, Fig. S5 C and D). Thus, WOBE437 inhibited 2-AG uptake, but not hydrolysis (Fig. 1G). In agreement with the hydrolase activity-based protein profiling (ABPP) in mouse brain homogenate (Fig. 1H) and classical radioactivity-based hydrolytic assays in other biological matrices (SI Appendix, Fig. S4 C and D), WOBE437 did not inhibit the 2-AG hydrolyzing enzymes MAGL, ABHD6, and ABHD12. Importantly, WOBE437 was completely stable in the presence of the main EC-degrading enzymes (SI Appendix, Fig. S6A). Furthermore, it did not inhibit COX-2 activity and exhibited no binding to fatty acid binding protein 5 (FABP5) (SI Appendix, Fig. S6 CE). Because AEA and 2-AG transport was previously suggested to be bidirectional (15, 21), we also explored the effect of WOBE437 on EC efflux and release in U937 cells. WOBE437 moderately, but significantly, inhibited the efflux of AEA from preloaded cells (SI Appendix, Fig. S7 A and B) and blocked the release of both ECs, while increasing their intracellular levels in cells stimulated with ATP, thapsigargin, and ionomycin (SI Appendix, Fig. S7 C and D). We then treated cells with radiolabeled 14C-WOBE437 to investigate its cellular penetration. In U937 cells, the vast majority of the radioactive signal was collected in the extracellular and membrane-bound fractions (75–90%), with negligible amounts (5–9%) detected in the intracellular fraction (Fig. 2 AD and SI Appendix, Table S4). Interestingly, the highest membrane-associated signal was shown at the concentration of 10 nM (9.8% of total signal), whereas at higher 14C-WOBE437 concentrations, the membrane-associated radioactivity significantly dropped to 1.7%, 0.7%, and 0.7% of total signal for the concentration of 100 nM, 1 μM, and 10 μM, respectively (Fig. 2 AD and SI Appendix, Table S4). These data suggest a saturable membrane target for WOBE437 (Fig. 2 AD and SI Appendix, Table S4 and Fig. S8). To further investigate this point, 14C-WOBE437 was applied to membrane preparations generated from mouse brain and U937 cells. The data showed a clear saturation binding kinetics with similar Kd values (15.7 ± 2.8 nM and 11.3 ± 2.4 nM for membrane preparations derived from mouse brain and U937 cells, respectively) and Bmax values (18.2 ± 1.5 fmol/mg protein and 15.6 ± 1.4 fmol/mg protein for mouse brain and U937 membranes, respectively) (Fig. 2 E and F). The Kd values are in line with the IC50 values shown in the functional AEA uptake assays in Neuro2a and U937 cells (Fig. 1 C and D). In further experiments, 10 nM 14C-WOBE437 was coincubated with different concentrations of AEA and U937 membranes. Increasing amounts of AEA induced a right-down shift of the 14C-WOBE437 binding curve, leading to an increase of Kd values, while not significantly modifying Bmax values (Fig. 2F and SI Appendix, Table S5). Furthermore, the EC AEA and noladin ether and the AEA uptake inhibitor OMDM-2 could compete with 14C-WOBE437 binding in brain and U937 membranes (SI Appendix, Fig. S9). Together, these data indicate a saturable membrane binding site for WOBE437 that is competitive with ECs and the membrane transport inhibitor OMDM-2. In intact U937 cells, Michaelis–Menten analysis of AEA transport kinetics displayed a competitive type of inhibition (Fig. 2G and SI Appendix, Table S6). Secondary plots of the slope (Kmapp/Vmaxapp vs. [WOBE437]) and intercept (1/Vmaxapp vs. [WOBE437]) further confirmed the competitive inhibition of AEA uptake (SI Appendix, Fig. S10). In subsequent experiments, the mechanism of AEA uptake inhibition by WOBE437 was shown to be reversible (rapid dilution assay; SI Appendix, Fig. S11 A and B) and independent of the preincubation time (SI Appendix, Fig. S11 and C). Incubation of 14C-WOBE437 (100 nM) with U937 cells over time (0–30 min) did not show any increase of the cell-associated radioactive signal (∼10% of total signal; SI Appendix, Fig. S11D). The same result was obtained by coincubating 14C-WOBE437 with 50 times higher concentration of AEA, thus suggesting that WOBE437 does not act as a substrate of EC transport (SI Appendix, Fig. S11D).
Fig. 2.
In vitro characterization of WOBE437. (AD) Different initial amounts of 14C-WOBE437 were incubated for 15 min with 2 × 106 U937 cells at 37 °C. Afterward, cells were centrifuged, and the 14C-signal was measured from the extracellular (light gray), membrane-bound (dark gray), and intracellular (white) fractions. The results are reported in absolute amounts and as percentage of the initial amount (SI Appendix, Table S4). (E and F) Binding kinetics of 14C-WOBE437 using membrane preparations of mouse brain (E) and U937 cells (F). In U937 cell membranes, 14C-WOBE437 was coincubated with vehicle or different concentrations of AEA. (G) Competitive inhibition of AEA cellular uptake in U937 cells as shown by Michaelis–Menten analysis. Apparent Km and Vmax values are shown in SI Appendix, Table S6. Data shown are mean values ± SD from at least four independent experiments each performed in triplicate.
We further investigated cell penetration by quantifying intracellular levels of WOBE437 by LC-MS/MS and measuring its distribution in the parallel artificial membrane permeability assay (PAMPA) (SI Appendix, Fig. S12 and Table S7). Both results confirmed that WOBE437 did not significantly penetrate cell membranes up to 10 µM.
It is noteworthy that in a CEREP screen, WOBE437 did not exhibit any significant interaction with 45 CNS-related receptors, including cannabinoid receptors (SI Appendix, Fig. S13A), highlighting its selectivity toward EC membrane transport. A full assessment of the binding of WOBE437 to CB1 and CB2 receptors revealed only negligible binding interactions (Ki values of 17 and 48 µM, respectively) (SI Appendix, Fig. S13B and Table S8).

Pharmacological Effects of WOBE437 in Mice.

WOBE437 was initially assessed in a battery of four individual tests typically associated with CB1 receptor activation in mice (nociception, locomotion, body temperature, and cataleptic behavior), collectively referred to as the “tetrad” (26). Dose–response experiments identified 10 mg/kg as the lowest dose to elicit a moderate, but complete, tetrad in BALB/c mice upon i.p. administration (Fig. 3 AD). Compared with the CB1 receptor agonist (R)-WIN55,212-2, the observed hypothermia, catalepsy, and antinociception were significantly less pronounced, whereas the reduction in locomotion was comparable (SI Appendix, Fig. S14 AD). To assess whether the WOBE437-induced effects were CB1 receptor-mediated, mice were pretreated with the selective antagonist/inverse agonist rimonabant (SR1). As in the case of (R)-WIN55,212-2, the WOBE437-induced hypothermia, catalepsy, and antinociception were completely blocked by SR1, whereas hypolocomotion was partially reversed (Fig. 3 AD). The tetrad was fully replicated in wild-type C57BL6/N mice and completely absent in CB1 receptor-deficient littermates (SI Appendix, Fig. S14 EH). These experiments provided a proof-of-concept for the mechanism of action of WOBE437 as an indirect CB1 receptor agonist. In subsequent studies, we evaluated the effects of WOBE437 in pain, inflammatory, and anxiety models. At doses of 5 and 10 mg/kg, WOBE437 elicited significant analgesic and antiinflammatory effects, as indicated by the reduced number of abdominal stretches in the acetic writhing test, comparable with indomethacin (5 mg/kg), and by protective effects in lipopolysaccharide (LPS)-induced endotoxemia in BALB/c mice (Fig. 3 E and F and SI Appendix, Fig. S15 A and B). Both effects were reversed by SR1. In the formalin test, WOBE437 reduced inflammatory pain and paw thickness comparable with indomethacin (5 mg/kg) (Fig. 3 G and H and SI Appendix, Fig. S15C). At the lower dose of 3 mg/kg, WOBE437 still showed significant anxiolytic effects in C57BL6/N mice in the elevated plus maze (EPM) and in the holeboard (HB) test (Fig. 3 I and J). The number of entries and the time spent in the open arms of the EPM were both significantly increased compared with vehicle-treated animals (Fig. 3I and SI Appendix, Fig. S16 A and B). The anxiolytic effect was completely blocked by SR1, indicating an indirect CB1 receptor-mediated mechanism for WOBE437. The same dose of 3 mg/kg was ineffective in the tetrad test and did not alter locomotion in the EPM assay (SI Appendix, Fig. S16 C and D). In the HB test, the head-dipping frequency was significantly increased at 1 h after injection, thus reconfirming the anxiolytic effect of WOBE437 (Fig. 3J).
Fig. 3.
Pharmacological profile of WOBE437 in vivo. (AD) Concentration-dependent WOBE437-induced hypothermia, hypolocomotion, catalepsy, and analgesia (tetrad) in BALB/c mice. The tetrad was assessed after 1 h from WOBE437 injection (i.p.). Hypothermia, catalepsy, and analgesia were fully blocked by pretreating the animals for 30 min with SR1, whereas hypolocomotion was only partially reversed. (E) WOBE437 (10 mg/kg) reduced the number of abdominal stretches in the acetic acid-induced writhing test. This analgesic effect was prevented by pretreatment with SR1. (F) WOBE437 (5 mg/kg) exerted protective effects in the LPS-induced drop of rectal temperature, which is a consequence of endotoxin challenge. The protective effect was inhibited by pretreatment with SR1. (G and H) Analgesic and antiinflammatory effects of WOBE437 during the second phase of formalin-induced pain and inflammatory responses. For E, G, and H, indomethacin (Indometh.) (5 mg/kg) is shown as reference analgesic drug. (I and J) Anxiolytic effects of WOBE437 in the EPM (I) and the HB (J) tests performed in C57BL6/N mice. The data show mean values ± SEM. Groups were compared with the vehicle treated control group or as indicated by arcs using a one-way (AE and GI) or two-way (F) ANOVA after Bonferroni's post hoc test or unpaired t test (GJ). n = 5–20 mice per group. *P < 0.05; **P < 0.01; ***P < 0.001. ns, not significant. All doses are indicated in mg/kg, i.p. Indometh., indomethacin.

WOBE437 Selectively Modulates 2-AG and AEA Concentrations in Mice.

To investigate whether there was a link between the in vivo pharmacological effects of WOBE437 and the inhibition of EC uptake in vitro, the levels of WOBE437, AEA, 2-AG, and related lipids in different tissues were quantified by LC-MS/MS. After i.p. injection of 10 mg/kg WOBE437, the peak plasma concentration of 492 ± 103 pmol/mL was reached after 15 min and decreased with a half-life of 203 min (Fig. 4A and SI Appendix, Fig. S17A). In the brain, the highest concentration was measured after 30 min (919 ± 314 pmol/g brain), remaining unchanged up to 1 h (652 ± 165 pmol/g brain). The Cbrain/Cplasma ratio (Kp) of WOBE437 was 0.65 after 5 min, 1.64 after 15 min, and reached 1.9 after 30 and 60 min (Fig. 4 A and E). The peak concentration of WOBE437 in the brain was 780 ± 267 nM, thus being in line with the bioactive concentrations determined in cellular assays. WOBE437 did not inhibit any of the major serine hydrolases in vivo (SI Appendix, Fig. S17 BD). After single and repeated injection(s), AEA and 2-AG levels in the plasma were not significantly affected (Fig. 4 B and C). Only at 15 min after injection, WOBE437 prevented the moderate increase of 2-AG induced by DMSO (Fig. 4C). Interestingly, although WOBE437 did not alter corticosterone levels after a single injection, it doubled the circulating concentration of corticosterone after 7 d (Fig. 4D). In the total brain, AEA levels did not change compared with basal level after a single injection with WOBE437 (Fig. 4F). On the contrary, 2-AG levels increased 1.7-fold compared with vehicle at 1 h after injection (Fig. 4G). After 7 d of daily treatment, both AEA and 2-AG levels were significantly raised by 1.5-fold compared with DMSO. The moderate, but significant, increase of both ECs in total brain did not affect the number of functional CB1 receptors (SI Appendix, Fig. S18). Repeated administrations of WOBE437 led to a significant threefold increase of corticosterone in the brain, which mirrors the changes observed in plasma (Fig. 4H). In peripheral organs, WOBE437 reached peak concentrations of 5–20 nmol/g tissue, without any signs of accumulation over 7 d (SI Appendix, Figs. S21–S23). After a single injection, the AEA level did not change in liver, whereas in kidney and spleen, it moderately diminished (SI Appendix, Figs. S21–S23). Similarly, 2-AG levels were not altered in the spleen, but dropped significantly after 60–360 min in liver and kidney (SI Appendix, Figs. S21–S23). However, EC concentrations were not affected by repeated administrations of WOBE437 in peripheral tissues. Again, corticosterone levels were increased in spleen, liver, and kidney (SI Appendix, Figs. S21–S23). Importantly, the concentrations of other lipids (NAEs, AA, prostaglandins, and progesterone) were not significantly altered upon WOBE437 treatment in brain, plasma, and peripheral organs (SI Appendix, Figs. S19–S23). We also determined the tissue distribution of 14C-WOBE437 (10 mg/kg, i.p.) in C57BL6/N mice after 1 h, and the highest 14C signal was measured in adipose tissue, followed by the spleen, liver, and kidney (SI Appendix, Fig. S24). The LC-MS/MS and 14C isotope quantifications provided a comparable tissue distribution of WOBE437, indicating its CNS bioavailability.
Fig. 4.
LC-MS/MS quantification of WOBE437, AEA, 2-AG and corticosterone in brain and plasma of C57BL6/N mice treated with 10 mg/kg of WOBE437 for different times. Plasma (AD) and brain (EH) concentrations of WOBE437, AEA, 2-AG, and corticosterone measured at different time points after treatment with 10 mg/kg (i.p., single injection and repeated administrations, once daily for 7 d) in C57BL6/N mice. Data represent means ± SD; n = 5 mice per group. Statistical analysis was performed by using one-way ANOVA to compare distinct groups of animals injected with DMSO or WOBE437 and killed at different time points. At every time point, DMSO- and WOBE437-treated animals were compared by using the two-tailed unpaired t test. *P < 0.05; **P < 0.01; ***P < 0.001, WOBE437 vs. vehicle; #P < 0.05; ##P < 0.01 vs. baseline (time 0); +P < 0.05; ++P < 0.01 repeated administration vs. single administration (at 60 min after injection).

Development of the Irreversible EC Transport Inhibitor RX-055.

Based on the WOBE437 chemical scaffold, we investigated the possibility of designing a tailored functional photoaffinity probe. Analogs were prepared by replacing either of the two methoxy groups on the phenyl ring with a hexyloxy chain bearing a photoactivatable trifluoromethylphenyl diazirine moiety (SI Appendix, Fig. S25). Intriguingly, the ability of these probes to inhibit AEA uptake was strongly dependent on the positioning of the linker between the phenyl moiety of the WOBE scaffold and the photoactivatable head group. Thus, 2 (RX-055), which has the hexyloxy linker in the position meta to the amide nitrogen, retained the same potency as WOBE437 (IC50 = 14 ± 3 nM, Hill slope = −0.791; Fig. 5 A and B and SI Appendix, Table S9), whereas the parasubstituted derivative 48 was a significantly weaker inhibitor of AEA uptake, with an IC50 value of 1,979 nM (SI Appendix, Fig. S26 A and B). Similar to WOBE437, RX-055 (500 nM) inhibited AEA uptake in U937 cells independently of preincubation time (SI Appendix, Fig. S11C). In Neuro2a cells, RX-055 (1 μM) inhibited the overall AEA uptake by 30–40% in the time range 2–20 min (Fig. 5C).
Fig. 5.
In vitro pharmacological characterization of the tailored irreversible EC transport inhibitor RX-055. (A) UV-induced mechanism of photoactivation of the diazirine group. (B) Inhibition curve (normalized) of AEA and 2-AG uptake in U937 cells for RX-055. (C and D) Time course of AEA (C) and 2-AG (D) uptake in adherent Neuro2a cells in presence of RX-055 (0.5 μM) or vehicle. (E and F) AEA (E) and 2-AG (F) levels measured by LC-MS/MS in isolated human whole blood after 1 h stimulation with LPS (2 μg/mL) and 15 min pretreatment with WOBE437 (1 μM), RX-055 (1 μM), or vehicle. The open circles are basal control (not LPS-challenged) and the closed circles are LPS-challenged. Data show mean values ± SEM (AD) calculated from at least three independent experiments each performed in triplicate. (E and F) Whole blood was collected from 10–12 different human donors. Statistical analysis was carried out by using unpaired two-tailed t test (BD) and a one-way ANOVA following Dunnett’s post hoc test (E and F). *P < 0.05; **P < 0.01; ***P < 0.001.
RX-055 also inhibited 2-AG uptake with low nanomolar potency (IC50 = 32 ± 7 nM, Hill slope = −0.694; Fig. 5B). In washout experiments, the inhibition of AEA and 2-AG uptake was maintained for RX-055 as opposed to WOBE437, for which AEA inhibition was significantly reduced (SI Appendix, Fig. S26 CE). RX-055 (0.5 μM) was also more efficient than WOBE437 in inhibiting 2-AG uptake in adherent Neuro2a cells (Fig. 5D). These data clearly indicate that, upon photoactivation, RX-055 covalently binds to a membrane target involved in AEA and 2-AG uptake. The probe was profiled on the main ECS targets showing no significant binding to CB1 and CB2 receptors (Ki > 100 μM) and negligible effects on MAGL, ABHD6, and ABHD12 up to 10 μM and an IC50 value for FAAH of 4 μM (SI Appendix, Fig. S27), corresponding to a 250-fold selectivity for AEA uptake inhibition. Similarly to WOBE437, RX-055 moderately inhibited the efflux of AEA and 2-AG from preloaded U937 cells and blocked the release of both ECs in stimulated cells (SI Appendix, Fig. S28).
Next, we investigated the effects of RX-055 (1 μM) on EC levels in isolated human whole blood challenged with LPS for 1 h. As shown in Fig. 5 E and F, the incubation with LPS raised the plasma concentrations of AEA and 2-AG, and pretreatment with RX-055 further increased the levels of both ECs by 30–40%. In the same experiments, 1 μM WOBE437 qualitatively behaved like RX-055, but with a lower efficacy (Fig. 5 E and F). It is noteworthy that both inhibitors did not affect the plasma concentrations of other NAEs (SI Appendix, Fig. S29). To explore the potential utility of this probe in vivo, RX-055 was assessed in an animal model of anxiety behavior. Because of limited systemic bioavailability, RX-055 was injected into specific brain regions and UV-photoactivated in situ by using cannula implantations temporarily inserted by a fiber-optic cable (Fig. 6A). At 90 min after treatment, the EPM test revealed that RX-055–injected mice showed significantly increased numbers of entries and time spent in the open arms compared with vehicle-injected and UV-irradiated littermate control mice (Fig. 6 B and C). The anxiolytic effects were evident at all tested concentrations injected bilaterally into the basolateral amygdala (BLA), whereas they were completely absent upon injection into the lateral ventricle (LV) (Fig. 6 B and C). RX-055 did not affect locomotion in the animals (SI Appendix, Fig. S30).
Fig. 6.
Anxiolytic effects of RX-055 upon injection and in situ photoactivation in the BLA in C57BL6/J mice. (A) Schematic representation of intracranial injection of RX-055 followed by in situ UV photoactivation in the BLA. (B and C) At all doses tested, RX-055 increased the open arm entries (B) and the time spent in open arms (C) upon injection and photoactivation in the BLA. In contrast, no significant changes of the same parameters were observed upon injection in the LV. In B and C, the control group was injected with solvent (DMSO) and UV-irradiated similarly to the RX-055 injected mice. The data show mean values ± SEM. Groups were compared with the vehicle-treated control group or as indicated by arcs using a one-way ANOVA following Bonferroni's post hoc test. n = 12–36 mice per group. *P < 0.05; **P < 0.01. ns, not significant.


The development of tool compounds that potently and selectively modulate the different components of the ECS has been fundamental in the exploration of the biochemical basis and pharmacological potential of this complex signaling network (11, 12, 27, 28). However, the lack of selective inhibitors of EC membrane transport has prevented an unambiguous biochemical and pharmacological characterization of this process. Hence, the exact EC transport mechanism remains elusive (21). Here, we report the design and generation of the highly potent and selective EC reuptake inhibitors WOBE437 (1) and RX-055 (2). WOBE437 inhibits AEA and 2-AG cellular uptake at substoichiometric concentrations (relative to the ECs) without altering the levels of other NAEs in cell lines and human whole blood, thus serving as a selective EC uptake inhibitor (for comparison with known AEA uptake inhibitors, see SI Appendix, Table S2). Intriguingly, despite the fact that WOBE437 can inhibit both EC release and reuptake, the net effect observed in complex systems (i.e., whole blood and mice) showed an increase of EC levels. A possible explanation for this finding may derive from a higher affinity of WOBE437 for reuptake inhibition, which could be due to its limited cell penetration. Indeed, residual AEA and 2-AG reuptake upon WOBE437 treatment is in the range of 20–50% (depending on the cell line), whereas it can inhibit only 20% of EC release (i.e., 80% residual release) (SI Appendix, Fig. S7). Moreover, in tissues, the catabolic and metabolic EC enzymes driving the release and facilitated uptake may be expressed in different cell types. Binding kinetics further indicate that WOBE437 binds to a membrane site with high affinity (10–20 nM). However, a second, low-affinity binding site and/or nonspecific binding to phospholipids cannot be excluded. WOBE437 is hydrolytically stable, and after i.p. administration at 10 mg/kg, it rapidly and efficiently accumulates in the brain (Kp > 1 after 15 min). To obtain biological insights into the potential role of EC uptake in vivo, we thoroughly characterized the pharmacological and biochemical effects of WOBE437. In mice, 10 mg/kg WOBE437 elicited a full tetrad response [although less pronounced than (R)-WIN55,212-], which is a hallmark of either direct CB1 receptor activation or the simultaneous elevation of AEA and 2-AG levels in the brain (Fig. 3 AD) (26, 29). In contrast, FAAH inhibitors only trigger analgesia, whereas MAGL inhibitors induce hypolocomotion, hypothermia, and analgesia, but not catalepsy (11, 12). The LC-MS/MS data showed that, in total brain tissue, acute WOBE437 treatment significantly increased 2-AG levels without affecting AEA (Fig. 4 F and G). Although the CB1 receptor-mediated WOBE437 induced behavioral changes might suggest that the significant increase in 2-AG was likely accompanied by a mild and/or region-specific rise of AEA levels that was not detectable, it cannot be ruled out that 2-AG uptake inhibition might differ pharmacologically from MAGL inhibition. At 5–10 mg/kg, WOBE437 elicited significant analgesic and antiinflammatory effects in different animal models; at the subtetrad-inducing dose of 3 mg/kg, the inhibitor exhibited anxiolytic effects in two mouse models of anxiety behavior (Fig. 3 I and J), similar to FAAH and MAGL inhibitors (11, 14, 29, 30). It is noteworthy that, after 7 d of treatment, both AEA and 2-AG levels had significantly increased in total brain by a factor of ∼1.5 compared with vehicle. Similarly, WOBE437 accumulated in the brain, reaching an estimated concentration of 926 nM after 7 d of treatment (10 mg/kg i.p., daily) compared with 555 nM 1 h after a single injection (Fig. 4 A and E). The levels of other NAEs remained essentially unchanged, suggesting that FAAH activity was not affected, which is also in agreement with the minor cell penetration of WOBE437 (Fig. 2 AD and SI Appendix, Fig. S12). The moderate increase of AEA and 2-AG concentrations induced by WOBE437 did not alter the number of functional CB1 receptors in the brain (SI Appendix, Fig. S20). In contrast, the prolonged 2-AG “overflow” (10–12 times basal levels) resulting from repeated administrations of JZL184 has been shown to desensitize CB1 receptors (31). However, repeated low doses of JZL184, in combination with FAAH inhibition, produce antinociceptive and anxiolytic effects without causing CB1 functional antagonism (32, 33). Similarly, a reversible MAGL inhibitor showed protective effects in a mouse model of multiple sclerosis after 21 d of treatment, which was associated with a twofold increase of 2-AG levels in the brain (34). Our data show that a selective competitive EC reuptake inhibitor modulates the homeostasis of AEA and 2-AG levels in a time- and space-restricted manner, without leading to an EC overflow or altering the levels of other lipids.
Intriguingly, after repeated administrations, WOBE437 (10 mg/kg) did not significantly alter the levels of AEA, 2-AG, and NAEs in kidney and liver (SI Appendix, Figs. S21 and S22). These data represent another pharmacological difference between the inhibition of EC reuptake and the blockage of EC degradation. The possibility to tissue-specifically increase EC levels provides a targeted approach without leading to chronic activation of liver and kidney CB1 receptors that exacerbate inflammation, promoting liver and renal fibrosis, insulin resistance, steatosis, and nephropathy (3538).
Unexpectedly, upon repeated administration, WOBE437 led to a significant twofold to fourfold increase of the corticosterone levels in brain, plasma, and peripheral organs (Fig. 4 D and H and SI Appendix, Figs. S19–S23). Although this observation requires further investigations, the modulation of corticosterone (the rodent equivalent of cortisol) tissue levels is interesting. A potential correlation between Δ9-THC–induced suppression of neuroinflammation and activation of the hypothalamic–pituitary–adrenal (HPA) axis in multiple sclerosis was recently discussed (39). Although stress conditions are usually characterized by high cortisol levels, burnout patients have an impaired response of the HPA axis, which leads to hypocortisolism (40, 41). In these patients, the chronic low level of cortisol has been correlated with the severity of clinical and nonclinical symptoms (42, 43).
To further study the mechanism of EC reuptake inhibition, we designed the suitable photoaffinity probe RX-055. This probe irreversibly blocked the uptake of both AEA and 2-AG with essentially the same potency as WOBE437, thus offering opportunities to elucidate the mechanism of EC reuptake and to investigate the pharmacology of its irreversible blockage. Administration of both WOBE437 and RX-055 to human whole blood and subsequent LPS stimulation showed a specific increase of ECs (Fig. 5 E and F), in line with the in vitro effects of these compounds and the cannabimimetic in vivo pharmacology observed with WOBE437. Generally, RX-055 was more potent and produced longer-lasting effects than WOBE437 on 2-AG uptake in vitro. These data are in agreement with a competitive reversible mechanism of action (WOBE437) that is dependent on the amount of substrate (100 nM AEA vs. 1 μM 2-AG) vs. an irreversible blockage (RX-055). When RX-055 was injected in the BLA, followed by in situ UV irradiation, it showed evident anxiolytic effects, whereas no effects were observed upon injection into the LV (Fig. 6). The role of BLA in cannabinoid-mediated anxiolytic effects has been shown (5, 9, 44). The difference between BLA and LV may reflect a distinct distribution of the probe and/or its inability to diffuse out from the LV into other brain regions.
Overall, these data show that EC membrane trafficking and cellular reuptake can be targeted. The potent and selective blockage of EC membrane transport by WOBE437 and RX-055 thus represents an alternative type of pharmacological modulation of the ECS. As shown, the currently known cytoplasmic AEA shuttling proteins (cytoplasmic transporters) like FABP5, hsp70, and FLAT1, or TRPV1, which was postulated to mediate AEA uptake in endothelial cells (21, 45), are not the target of WOBE437. Therefore, the present study strongly indicates that EC membrane transport mechanisms independent from cytoplasmic binding proteins or degrading enzymes play a crucial role in the brain, as previously suggested from data with less selective and potent AEA reuptake inhibitors (21). Being selective low nanomolar EC transport inhibitors, WOBE437 and RX-055 are not only suitable tool compounds to study lipid membrane transport mechanisms, which is highly challenging, but also address the need for high-quality chemical probes for research (46). Moreover, RX-055 is a unique covalent probe to unravel the biochemistry of EC trafficking and could be instrumental for the identification of the membrane proteins involved in this process.

Materials and Methods


An extended section is provided in SI Appendix, SI Experimental Procedures. Animal experiments were approved by the Local Animal Welfare Committees, in strict compliance with the ethical guidelines of the European Union (University of Mainz, University of Bern, and Uppsala University) and Mexican Federal Regulations for the Care and Use of Laboratory Animals (University of Guadalajara).

Chemical Synthesis of N-Alkylamides and WOBE437.

The synthesis of compounds is described in the SI Appendix, SI Experimental Procedures.

Inhibition of AEA and 2-AG Uptake and Release in Different Cell Types.

All radioactivity-based AEA and 2-AG uptake and release assays performed in U937 and HMC-1 cells were carried out by using two different assay formats as reported (15, 25, 47). AEA and 2-AG uptake assay in Neuro2a and primary rat cortical neurons were performed with a different method described in SI Appendix, SI Experimental Procedures.


ABPP experiments were performed by using membrane preparations obtained from mouse brains as described in ref. 25 and in SI Appendix, SI Experimental Procedures.

Enzyme Activity and Metabolic Stability Assays.

FAAH, ABHDs, MAGL, and COX-2 activity assays were performed by using cell homogenates, human recombinant enzymes, and mouse, rat, and pig brain homogenates as described in refs. 15, 25, 47, and 48 and reported in SI Appendix, SI Experimental Procedures.

Binding Assay, CB1 Functional Assays and in Vitro Pharmacological Screening.

CB1 and CB2 receptor binding assays and [35S]GTPγS assays were performed as described in ref. 49 and reported in SI Appendix, SI Experimental Procedures. WOBE437 was profiled in more detail in an extended CNS adverse drug reaction screening panel (CEREP) (50). The list of targets is reported in SI Appendix, SI Experimental Procedures.

Localization and Quantification of 14C-WOBE437 in Cellular Fractions of U937 Cells and Mouse Tissues.

The cell penetration and tissue distribution of 14C-WOBE437 was assessed in U937 cells and C57BL6/N mice as described in SI Appendix, SI Experimental Procedures.

In Vivo Experiments in BALB/c, C57BL6 and Phenotypic CB1-KO Mice.

WOBE437 and RX-055 were tested in independent laboratories, in different mouse strains, and in multiple (behavioral) models for cannabimimetic, analgesic, antiinflammatory, and anxiolytic effects in wild-type and phenotypic CB1-KO mice as described in SI Appendix, SI Experimental Procedures.

LC-MS/MS Quantification of WOBE437, ECs, and Other Lipids.

WOBE437, AEA, 2-AG, and other lipids and hormones were quantified by LC-MS/MS in different biological matrices (cells, human whole blood, and rodent tissues) as reported in SI Appendix, SI Experimental Procedures.


We thank M. Salome Gachet for measuring the UCM707 and WOBE437 distribution in U937 cells and Mark Rau, Sabine Rihs, Patricia Schenker, and Tatiana Hofer for performing some replicate measurements. This work was supported by the Swiss National Science Foundation NCCR TransCure and the EIN Roche grant. F. Hoffmann-La Roche provided support for the PAMPA assay and in vitro profiling and confirmation of CB receptor binding data. M.B. was supported by the Brazilian Research Council and the Fredrik, Ingrid Thurings Stiftelse and Stiftelse Lars Hiertas Minne and Åhlén-stiftelse. M.S. and C.A. were supported by Bundesministerium für Wirtschaft und Energie (ZIM KOOP) Grant KF2611301MD0.

Supporting Information

Appendix (PDF)


M Maccarrone, et al., Endocannabinoid signaling at the periphery: 50 years after THC. Trends Pharmacol Sci 36, 277–296 (2015).
M Maccarrone, M Guzmán, K Mackie, P Doherty, T Harkany, Programming of neural cells by (endo)cannabinoids: From physiological rules to emerging therapies. Nat Rev Neurosci 15, 786–801 (2014).
D Piomelli, O Sasso, Peripheral gating of pain signals by endogenous lipid mediators. Nat Neurosci 17, 164–174 (2014).
R Mechoulam, LA Parker, The endocannabinoid system and the brain. Annu Rev Psychol 64, 21–47 (2013).
S Ruehle, AA Rey, F Remmers, B Lutz, The endocannabinoid system in anxiety, fear memory and habituation. J Psychopharmacol 26, 23–39 (2012).
T Rubino, E Zamberletti, D Parolaro, Endocannabinoids and mental disorders. Handb Exp Pharmacol 231, 261–283 (2015).
V Di Marzo, M Bifulco, L De Petrocellis, The endocannabinoid system and its therapeutic exploitation. Nat Rev Drug Discov 3, 771–784 (2004).
V Di Marzo, Targeting the endocannabinoid system: To enhance or reduce? Nat Rev Drug Discov 7, 438–455 (2008).
B Lutz, G Marsicano, R Maldonado, CJ Hillard, The endocannabinoid system in guarding against fear, anxiety and stress. Nat Rev Neurosci 16, 705–718 (2015).
LA Batista, PH Gobira, TG Viana, DC Aguiar, FA Moreira, Inhibition of endocannabinoid neuronal uptake and hydrolysis as strategies for developing anxiolytic drugs. Behav Pharmacol 25, 425–433 (2014).
S Kathuria, et al., Modulation of anxiety through blockade of anandamide hydrolysis. Nat Med 9, 76–81 (2003).
JZ Long, et al., Selective blockade of 2-arachidonoylglycerol hydrolysis produces cannabinoid behavioral effects. Nat Chem Biol 5, 37–44 (2009).
FA Moreira, N Kaiser, K Monory, B Lutz, Reduced anxiety-like behaviour induced by genetic and pharmacological inhibition of the endocannabinoid-degrading enzyme fatty acid amide hydrolase (FAAH) is mediated by CB1 receptors. Neuropharmacology 54, 141–150 (2008).
SG Kinsey, ST O’Neal, JZ Long, BF Cravatt, AH Lichtman, Inhibition of endocannabinoid catabolic enzymes elicits anxiolytic-like effects in the marble burying assay. Pharmacol Biochem Behav 98, 21–27 (2011).
A Chicca, J Marazzi, S Nicolussi, J Gertsch, Evidence for bidirectional endocannabinoid transport across cell membranes. J Biol Chem 287, 34660–34682 (2012).
M Beltramo, et al., Functional role of high-affinity anandamide transport, as revealed by selective inhibition. Science 277, 1094–1097 (1997).
CJ Hillard, WS Edgemond, A Jarrahian, WB Campbell, Accumulation of N-arachidonoylethanolamine (anandamide) into cerebellar granule cells occurs via facilitated diffusion. J Neurochem 69, 631–638 (1997).
M Beltramo, D Piomelli, Carrier-mediated transport and enzymatic hydrolysis of the endogenous cannabinoid 2-arachidonylglycerol. Neuroreport 11, 1231–1235 (2000).
D Fegley, et al., Anandamide transport is independent of fatty-acid amide hydrolase activity and is blocked by the hydrolysis-resistant inhibitor AM1172. Proc Natl Acad Sci USA 101, 8756–8761 (2004).
JL Blankman, BF Cravatt, Chemical probes of endocannabinoid metabolism. Pharmacol Rev 65, 849–871 (2013).
S Nicolussi, J Gertsch, Endocannabinoid transport revisited. Vitam Horm 98, 441–485 (2015).
S Raduner, et al., Alkylamides from Echinacea are a new class of cannabinomimetics. Cannabinoid type 2 receptor-dependent and -independent immunomodulatory effects. J Biol Chem 281, 14192–14206 (2006).
A Chicca, et al., Synergistic immunomopharmacological effects of N-alkylamides in Echinacea purpurea herbal extracts. Int Immunopharmacol 9, 850–858 (2009).
T Bisogno, et al., N-acyl-dopamines: Novel synthetic CB(1) cannabinoid-receptor ligands and inhibitors of anandamide inactivation with cannabimimetic activity in vitro and in vivo. Biochem J 351, 817–824 (2000).
S Nicolussi, et al., Guineensine is a novel inhibitor of endocannabinoid uptake showing cannabimimetic behavioral effects in BALB/c mice. Pharmacol Res 80, 52–65 (2014).
K Monory, et al., Genetic dissection of behavioural and autonomic effects of Δ(9)-tetrahydrocannabinol in mice. PLoS Biol 5, e269 (2007).
A Chicca, C Arena, C Manera, Beyond the direct activation of cannabinoid receptors: New strategies to modulate the endocannabinoid system in CNS-related diseases. Recent Patents CNS Drug Discov 10, 122–141 (2016).
F Piscitelli, V Di Marzo, “Redundancy” of endocannabinoid inactivation: New challenges and opportunities for pain control. ACS Chem Neurosci 3, 356–363 (2012).
JZ Long, et al., Dual blockade of FAAH and MAGL identifies behavioral processes regulated by endocannabinoid crosstalk in vivo. Proc Natl Acad Sci USA 106, 20270–20275 (2009).
NR Sciolino, W Zhou, AG Hohmann, Enhancement of endocannabinoid signaling with JZL184, an inhibitor of the 2-arachidonoylglycerol hydrolyzing enzyme monoacylglycerol lipase, produces anxiolytic effects under conditions of high environmental aversiveness in rats. Pharmacol Res 64, 226–234 (2011).
JE Schlosburg, et al., Chronic monoacylglycerol lipase blockade causes functional antagonism of the endocannabinoid system. Nat Neurosci 13, 1113–1119 (2010).
SG Kinsey, et al., Repeated low-dose administration of the monoacylglycerol lipase inhibitor JZL184 retains cannabinoid receptor type 1-mediated antinociceptive and gastroprotective effects. J Pharmacol Exp Ther 345, 492–501 (2013).
S Ghosh, et al., Full fatty acid amide hydrolase inhibition combined with partial monoacylglycerol lipase inhibition: Augmented and sustained antinociceptive effects with reduced cannabimimetic side effects in mice. J Pharmacol Exp Ther 354, 111–120 (2015).
G Hernández-Torres, et al., A reversible and selective inhibitor of monoacylglycerol lipase ameliorates multiple sclerosis. Angew Chem Int Ed Engl 53, 13765–13770 (2014).
F Teixeira-Clerc, et al., CB1 cannabinoid receptor antagonism: A new strategy for the treatment of liver fibrosis. Nat Med 12, 671–676 (2006).
P Mukhopadhyay, et al., CB1 cannabinoid receptors promote oxidative/nitrosative stress, inflammation and cell death in a murine nephropathy model. Br J Pharmacol 160, 657–668 (2010).
T Jourdan, et al., Overactive cannabinoid 1 receptor in podocytes drives type 2 diabetic nephropathy. Proc Natl Acad Sci USA 111, E5420–E5428 (2014).
L Lecru, et al., Cannabinoid receptor 1 is a major mediator of renal fibrosis. Kidney Int 88, 72–84 (2015).
G Pryce, D Baker, Endocannabinoids in multiple sclerosis and amyotrophic lateral sclerosis. Handb Exp Pharmacol 231, 213–231 (2015).
W de Vente, JG van Amsterdam, M Olff, JH Kamphuis, PM Emmelkamp, Burnout is associated with reduced parasympathetic activity and reduced HPA axis responsiveness, predominantly in males. BioMed Res Int 2015, 431725 (2015).
C Heim, U Ehlert, DH Hellhammer, The potential role of hypocortisolism in the pathophysiology of stress-related bodily disorders. Psychoneuroendocrinology 25, 1–35 (2000).
AK Lennartsson, A Sjörs, P Währborg, T Ljung, IH Jonsdottir, Burnout and hypocortisolism—a matter of severity? A study on ACTH and cortisol responses to acute psychosocial stress. Front Psychiatry 6, 8 (2015).
BG Oosterholt, JH Maes, D Van der Linden, MJ Verbraak, MA Kompier, Burnout and cortisol: Evidence for a lower cortisol awakening response in both clinical and non-clinical burnout. J Psychosom Res 78, 445–451 (2015).
E Ganon-Elazar, I Akirav, Cannabinoid receptor activation in the basolateral amygdala blocks the effects of stress on the conditioning and extinction of inhibitory avoidance. J Neurosci 29, 11078–11088 (2009).
CJ Fowler, Transport of endocannabinoids across the plasma membrane and within the cell. FEBS J 280, 1895–1904 (2013).
CH Arrowsmith, et al., The promise and peril of chemical probes. Nat Chem Biol 11, 536–541 (2015).
S Nicolussi, et al., Correlating FAAH and anandamide cellular uptake inhibition using N-alkylcarbamate inhibitors: From ultrapotent to hyperpotent. Biochem Pharmacol 92, 669–689 (2014).
A Chicca, et al., Functionalization of β-caryophyllene generates novel polypharmacology in the endocannabinoid system. ACS Chem Biol 9, 1499–1507 (2014).
O Rosati, et al., One-pot heterogeneous synthesis of Δ(3)-tetrahydrocannabinol analogues and xanthenes showing differential binding to CB(1) and CB(2) receptors. Eur J Med Chem 85, 77–86 (2014).
J Bowes, et al., Reducing safety-related drug attrition: The use of in vitro pharmacological profiling. Nat Rev Drug Discov 11, 909–922 (2012).

Information & Authors


Published in

Go to Proceedings of the National Academy of Sciences
Go to Proceedings of the National Academy of Sciences
Proceedings of the National Academy of Sciences
Vol. 114 | No. 25
June 20, 2017
PubMed: 28584105


Submission history

Published online: June 5, 2017
Published in issue: June 20, 2017


  1. endocannabinoid reuptake
  2. 2-AG
  3. inhibitor
  4. endocannabinoid system
  5. lipid transport


We thank M. Salome Gachet for measuring the UCM707 and WOBE437 distribution in U937 cells and Mark Rau, Sabine Rihs, Patricia Schenker, and Tatiana Hofer for performing some replicate measurements. This work was supported by the Swiss National Science Foundation NCCR TransCure and the EIN Roche grant. F. Hoffmann-La Roche provided support for the PAMPA assay and in vitro profiling and confirmation of CB receptor binding data. M.B. was supported by the Brazilian Research Council and the Fredrik, Ingrid Thurings Stiftelse and Stiftelse Lars Hiertas Minne and Åhlén-stiftelse. M.S. and C.A. were supported by Bundesministerium für Wirtschaft und Energie (ZIM KOOP) Grant KF2611301MD0.


This article is a PNAS Direct Submission.



Andrea Chicca1
Institute of Biochemistry and Molecular Medicine, National Centre of Competence in Research NCCR TransCure, University of Bern, 3012 Bern, Switzerland;
Simon Nicolussi1
Institute of Biochemistry and Molecular Medicine, National Centre of Competence in Research NCCR TransCure, University of Bern, 3012 Bern, Switzerland;
Ruben Bartholomäus
Department of Chemistry and Applied Biosciences, Institute of Pharmaceutical Sciences, ETH Zurich, 8093 Zurich, Switzerland;
Martina Blunder
Department of Neuroscience, Biomedical Center, Uppsala University, 751 24 Uppsala, Sweden;
Brain Institute, Universidade Federal do Rio Grande do Norte, Natal 59056-450, Brazil;
Alejandro Aparisi Rey
Institute of Physiological Chemistry, University Medical Center of the Johannes Gutenberg University Mainz, D-55099 Mainz, Germany;
Vanessa Petrucci
Institute of Biochemistry and Molecular Medicine, National Centre of Competence in Research NCCR TransCure, University of Bern, 3012 Bern, Switzerland;
Ines del Carmen Reynoso-Moreno
Institute of Biochemistry and Molecular Medicine, National Centre of Competence in Research NCCR TransCure, University of Bern, 3012 Bern, Switzerland;
Centro Universitario de Ciencias Exactas e Ingenierías, University of Guadalajara, 44430 Guadalajara, Mexico;
Juan Manuel Viveros-Paredes
Centro Universitario de Ciencias Exactas e Ingenierías, University of Guadalajara, 44430 Guadalajara, Mexico;
Marianela Dalghi Gens
Institute of Biochemistry and Molecular Medicine, National Centre of Competence in Research NCCR TransCure, University of Bern, 3012 Bern, Switzerland;
Beat Lutz
Institute of Physiological Chemistry, University Medical Center of the Johannes Gutenberg University Mainz, D-55099 Mainz, Germany;
Helgi B. Schiöth
Department of Neuroscience, Biomedical Center, Uppsala University, 751 24 Uppsala, Sweden;
Michael Soeberdt
Dr. August Wolff GmbH & Co. KG Arzneimittel, 33611 Bielefeld, Germany
Christoph Abels
Dr. August Wolff GmbH & Co. KG Arzneimittel, 33611 Bielefeld, Germany
Roch-Philippe Charles
Institute of Biochemistry and Molecular Medicine, National Centre of Competence in Research NCCR TransCure, University of Bern, 3012 Bern, Switzerland;
Karl-Heinz Altmann
Department of Chemistry and Applied Biosciences, Institute of Pharmaceutical Sciences, ETH Zurich, 8093 Zurich, Switzerland;
Jürg Gertsch2 [email protected]
Institute of Biochemistry and Molecular Medicine, National Centre of Competence in Research NCCR TransCure, University of Bern, 3012 Bern, Switzerland;


To whom correspondence should be addressed. Email: [email protected].
Author contributions: A.C., S.N., M.B., B.L., K.-H.A., and J.G. designed research; A.C., S.N., R.B., M.B., A.A.R., V.P., I.d.C.R.-M., J.M.V.-P., and M.D.G. performed research; M.S., C.A., and R.-P.C. contributed new reagents/analytic tools; A.C., S.N., R.B., M.B., A.A.R., V.P., I.d.C.R.-M., J.M.V.-P., B.L., H.B.S., M.S., and J.G. analyzed data; and A.C., S.N., M.S., K.-H.A., and J.G. wrote the paper.
A.C. and S.N. contributed equally to this work.

Competing Interests

The authors declare no conflict of interest.

Metrics & Citations


Note: The article usage is presented with a three- to four-day delay and will update daily once available. Due to ths delay, usage data will not appear immediately following publication. Citation information is sourced from Crossref Cited-by service.

Citation statements



If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download.

Cited by


    View Options

    View options

    PDF format

    Download this article as a PDF file


    Get Access

    Login options

    Check if you have access through your login credentials or your institution to get full access on this article.

    Personal login Institutional Login

    Recommend to a librarian

    Recommend PNAS to a Librarian

    Purchase options

    Purchase this article to get full access to it.

    Single Article Purchase

    Chemical probes to potently and selectively inhibit endocannabinoid cellular reuptake
    Proceedings of the National Academy of Sciences
    • Vol. 114
    • No. 25
    • pp. 6413-6544







    Share article link

    Share on social media